The URL can be used to link to this page

Home My WebLink AboutNYS Fertilizer & Pesticide Demonstration Plot Sept 1983SOUTHOI. D DEMONSTRATION SITE NEW YORK STATE FERTILIZER AND PESTICIDE DEMONSTRATION PLOT SEPTEMBER 1983 Center for Environmental Research Hollister Hell Ithaca, NY 14853 SOUTHOLD DEMONSTRATION SITE New York State Fertilizer and Pesticide Demonstration Project August 1983 by Nancy M. Trautmann Keith S. Porter Henry B. F. Hughes Illustrated by Donna Curtin CENTER FOR ENVIRONMENTAL RESEARCH 468 Hollister Hall Cornell University Ithaca~ N.Y. 14853 Sponsorship This project was sponsored by the New York State Department of Environmental Conservation as part of the Fiscal Year 1981 Ground Water Management Program Fertilizer and Pesticide Demonstration Project. Funding was provided by the United States Environmental Protection Agency under Section Z08 of the Water Pollution Control Act of 1972. Cornell University provided matching funds for this grant. Acknowledgments Many people have contributed to the field work and review of technical material over the course of this project. Primary among these were Dale D. Moyer, Potato Agent for the Suffolk County Cooperative Extension Association, who provided many hours of assistance and advice, and Robert Schneck, from the N.Y.S. Department of Environmental Conservation, Region I, who provided direction for the project and critical review of each of our draft reports. In addition, the following people have contributed time and expertise to the Southold Demonstration Project: Citizen's Advisory Committee Frank Bear Bud Cybulski Richard Hilary Fred Lappe James Monsell Frank Murphy Ruth Oliva William Ruland Valeria Scopaz Shaw Lydia Tortora, Chairperson Cooperative Extension Dale Moyer Daniel Fricke William Sanok David Newton New York State Department of Environmental Conservation P~ober t Schneck C. Thomas Male, Wl Other Government Agencies Aldo Andreoli - Suffolk County Department of tlealth Services Tom Martin - Suffolk County Department of Health Services Sarah Meyland - State Commission on Water Resource Needs of L.L James Monsell - Greenport Superintendent of Utilities Frank Murphy - Southold Town Councilman Supervisor William Pell, I17, and Members of the Southold Town Board ii Center for Environmental Research James Holway James Pike Steven Pacenka Mary Jane Heather Donna Curtin Gilbert Levine Carin Rundle ivl argaret Neno Katherine Wilson Cornell University Tammo Steenhuis Ann Lemley Jeff Wagenet David Bouldin Joseph Sieczka Arthur Bing iii Table of Contents Acknowledgments List of Figures List of Tables I. Introduction Regional Context for this Study Study Area Description Hydrogeologic Description Southold's Ground-Water Problems Objectives and Limitations of This Study An Introduction to the Water and Land Resource Analysis System The WALRAS Process Interpretation of WALRAS Results WALRAS Applied to Southold Climate Data Soil Data Land Use Data Land Management Data Effects of Housing Development on Ground-Water Quality Lawn Fertilization Rates Housing Densities Cluster vs. Sprawl Potential Contamination by Organic Chemicals IV. Effects of Potato Farming on Ground-Water Quality Potato Fertilization Rates Split Applications Effects of Heavy Spring Rains Economic Consequences of Various Fertilization Practices Potato Pest Management V. Effects of Other Land Uses on Ground-Water Quality Existing Land Uses Possible Future Land Uses Page 1 1 Z 3 4 7 7 8 10 11 11 11 lZ 13 13 15 19 ZZ Z9 Z9 31 34 35 36 41 41 41 Table of Contents (continued) Aldicarb Contamination of Southold's Ground ~¥ater Aldicarb Use in Southold Depot Lane Transect Aldicarb Simulation VII. Pesticide Screening to Prevent Ground-Water Contamination The Budget Approach Building Mathematical Models into the Pesticide Registration Process Guiding Field Investigations Improving Budget Estimates During Field Work Conclusions and Recommendations IX. Postscript APPENDICES A. Potato Pest Management on Long Island B. Input Data and Assumptions C. Tnrf Nitrogen Simulations D. Potato Nitrogen Simulations E. Glossary and Abbreviations Page 47 47 47 47 55 55 56 63 71 List of Figures 1-1. II-1. H-Z. lTr-1. rrr-Z. HI-3. IH-4. 11I-5. m--6. HI-7. 11I-8. 11I-9. IV-1. IV-?,. Distribution of aldicarb contamination of ground water in Southold~ N.Y. Root zone, unsaturated and saturated zones Movement of nitrate to and through the ground water Nitrogen concentrations of recharge to ground water from various lawn fertilization rates The effect of housing density on nitrate concentrations leaching to ground water in Southuld~ N.Y. Simulated nitrate leaching concentrations from a range of housing densities and lawn fertilization rates. Subarea of the Town of Southold~ Long Island, N.Y. The effect of clustered development on the nitrate concentration of recharge water The effect of various clustering options on nitrate concentrations leaching to ground water for a subarea of Southold~ N.Y. Effect of land uses on ground-water quality Four Long Island communities tested for contamination of ground water by organic chemicals Housing density compared with organic chemical contamination of ground water in four Long Island communities The effects of potato fertilization rates on nitrate concentrations leaching to ground water for two soil types in Southold, N.Y. Results of field experiments carried out from 1956 to 1978 at the L.LH.R.L. on yields of Katahdin potatoes at various fertilization rates Comparison of nitrate leaching from heavy fertilization at planting vs. lighter fertillzation with split application Effects of split application of fertilizer on simulated nitrate concentrations leaching to ground water from potato fields in Southold~ N.Y. Effects of split applications of fertilizer on simulated tuber nitrogen content in $outhold~ N.Y. List of Figures (continued) IV-7. V-1. V-Z. V-3. V--4. VI-1. VI-Z. VI-3. VII-1. VH-Z. VII-3. A-~. A-3. Effects of heavy spring rain on nitrogen lost from potato fields grown on Riverhead and Haven soils. Changes in population and acreages of farm land and land used for potato production in the Town of Southold, N.Y. from 1930 to 1980 Land uses in the Depot Lane transect area, Southold, N.Y. Soil types in the Depot Lane trausect area, Southold, N.Y. Results of nitrate simulations in the Depot Lane transect area, Southold, N.Y. Nitrate concentrations in ground water along the Depot Lane transect, Southold, N.Y. Aldicarb concentrations measured in wells along the Depot Lane transect, Southold, N.Y. Areas used for aldicarb analysis along the Depot Lane transect, Southold, N.Y. Simulated aldicarb plumes along the Depot Lane transect, Southold, N.Y. Generic model of pesticide behavior in the root zone Integration of mathematical models into the pesticide evaluation process Sensitivity of annual aldicarb mass balances to decomposition and plant uptake parameters Relationships among impact limits, uncertainties, and decisions about pesticides for potential use in Southold, N.Y. Proposed review procedure for pesticides, assuming the null hypothesis that the pesticide will not violate ground-water criteria or standards An aphid on the leaf of a hybrid potato plant that produces a sticky black gum, trapping insects and causing them to starve to death. A potato showing the effects of the potato scab disease. Barnyard grass (Echinochloa Crus-Galli) a common weed in potato fields in Southold, N.Y. vii List of Figures (continued) A-4. A-5. C-1. C-Z. D-1. D-Z. Nutsedge (Cyperus Rotudus), a common weed in potato fields in Southold, N.Y. Bindweed (Convolvulus Arvensis), a weed commonly found along roadsides which also invades potato fields in Southold, N.Y. Comparison of field data with WALRAS simulations of nitrogen uptake by turf Twelve Pines study area Comparison of WALRAS simulation results with field measurements of the nitrogen content of potato tubers. The field experiments were carried out by the Long Island Horticultural Research Laboratory Sensitivity analysis for the WALRAS potato simulations. viii List of Tables IV-3. V-1. VI-3 VII-1. The relationship between annual average nitrate concentrations and the probability that the standard will not be violated by individual samples New York State standards for water quality for contaminants considered in this report Rates and dates of lawn fertilization used in WALRAS simulations Assumptions related to nitrogen inputs for WALRAS simulations of residential development Comparison of Southold housing densities with the average percentage of time that the 10 rog/1 nitrate standard can be expected to be exceeded Description of organic chemicals found in ground water underlying residential communities on Long Island, New York Number of wells affected by organic chemical contamination in Wading River, Rocky Point, Mastic and Mastic Beach Comparison of average housing density with organic chemical and nitrogen contamination of ground water in four Long Island communities Simulated values for nitrogen leached and nitrogen content of potatoes grown on Riverhead and Haven soils. Economic consideration of potato fertilization rates. Per acre yield of potatoes on L.I. and in U.S., 1955-1980 Simulated nitrate leaching concentrations for various land uses in Southold, N.Y. Aldicarb use on fields along the Depot Lane transect, 1975 to 1979 Simulation results showing dates that aldicarb will reach ground water, and its average concentrations in ground water below the Depot Lane transect in Southold, N.Y. Laboratory data on half-lifes (in days) of aldicarb and its degradation products in water at various pH's and temperatures. Budget corresponding to model in Figure VII-1 B-Z. B-3. m-4o B-7o C-3. C-4. C-$. C-6o List of Tables (continued) Minimal data requirements for WALRAS regional evaluation of pesticide movement and fate Summary of assumptions relating to each land use type. Summary of assumptions relating to land cover types. Rates and timings of organic fertilizer (kg N/ha), Southold Demonstration Site. Rates and timings of organic slow-release fertilizer (kg N/ha), Southold Demonstration Site. Irrigation amounts and timings, Southold Demonstration Site. Nitrogen uptake curves (expressed as percentages), Southold Demonstration Site. Plant harvest fractions and dates (expressed as percentages), Southold Demonstration Site. Rates and dates of fertilizer N applied to turf (Selleck and others, 198o) Basic data for the Twelve Pines subdivision Simulated nitrogen in recharge, Twelve Pines area Summary of Twelve Pines well data Comparison of observed and simulated nitrogen concentrations in ground water, Twelve Pines subdivision Summary of sensitivity analysis of the WALRAS root zone water and nitrogen simulations for turf for conditions representing Southampton, N.Y. Fertilization rates and timings for L.I.H.R.L. experiments on Katahdin potatoes, 1972-1975. Comparison of field and simulation (WALRAS) results for nitrogen content of harvested potatoes X Chapter I I~TRODUCTION Regional Context for Tb~ Study Ground water is the primary source of drinking water for approximately half the population of the United States. On Long Island, ground water percolates through the soil and is stored in permeable geological formations called aquifers. The Long Island aquifer system is one of only seven in the country that has been designated by the U.S. Environmental Protection Agency as "sole source,~ and it provides one hundred percent of the drinking water for the 1.3 million residents of Suffolk County, an area encompassing the eastern two-thirds of the Island. Traditionally, it has been assumed that ground-water quality is protected by the filtering action of soil as the water percolates downward through the ground. In recent years, however, widespread discoveries of ground-water contam- ination have shown this assumption to be unfounded. Housing, industry, agriculture, and commercial land uses can pollute underlying ground water, as can more obvious sources such as landfills or toxic waste dumps. This report addresses ground-water contamination problems that have been identified in the Town of Southold, at the eastern end of Long Island, New York. Although Southold relies exclusively on ground water for its drinking water supplies, many of the area's drinking water wells exceed standards for nitrate or guideline values for certain agricultural pesticides. Synthetic organic chemicals used in industrial, commercial, and household products also have been detected in some of the Town's wells. The key to restoring and protecting the quality of Southold's aquifer is prevention of contamination of the recharge water that continuously replenishes the ground water supply. It is for this reason that decisions regarding the many land uses and human activities in Southold need to be made based on an understanding of how they will affect the quality of ground- water recharge supplies. In order to demonstrate basic procedures of ground-water resource management, the New York State Department of Environmental Conservation and the U.S. Environmental Protection Agency funded Cornell University's Center for Environmental Research to work with four New York communities on ways of protecting ground-water recharge from nitrate and pesticide pollution. This report describes the work done in Southold, N.Y., evaluating existing contamination sourdSs and means of reducing their impact on ground water. Although the results presented here are specific to the Southold site, the technical methods used can also be applied in other areas with similo, r ground-water concerns. The Southold Study Area Southold is a scenic rural town encompassing the North Fork at the eastern end of Long Island, N.Y. Bounded on three sides by salt water, it is characterized by coastal views and large stretches of land devoted to farming. Agriculture, fishing, and support of seasonal residents and tourists form the basis for the town's economy. Southold's year-round population was 19,000 in 1980, up 14 percent from a decade earlier, and the seasonal population has been expanding as well. The effects of this increased residential development on the town's open space amenities and ground- water quality are issues of growing concern in Southold. Ever since the invention of the mechanical potato digger in the 1880's, potatoes have been Southold's major crop, with approximately 5,500 acres currently devoted to potato fields. Close to one quarter of Long Island's potato crop is grown in Southold, where potato farming is a ten million dollar per year industry. P~elated industries including sales of agricultural equipment and supplies also are important to the North Fork economy. The monoculture nature of potato farming in Southold has led to severe pressure by the Colorado potato beetle, which has developed resistance over the years to many of the chemicals used for its control. In 1975, a new pesticide called aldicarb (brand name Temik) proved to 'be highly effective in controlling the beetle populations. This chemical was used by Southold's potato farmers through 1979, when it was withdrawn from sale on Long Island because of its discovery in the area's ground water. Because of the current difficulty in combating the Colorado potato beetle and the marginal economics of potato farming in recent years, many Southold farmers have diversified their crops or are contemplating doing so in the future. Cabbage, cauliflower, and sweet corn are the most common vegetable crops, and occasionally these are grown in rotation with potatoes. Although not a major percentage of the agricultural market, fruits including peaches, apples, pears, table grapes, cherries, melons, and berries also are grown in Southold. The area's newest agricultural crop is wine grapes, with the first vineyard started in 1973. The substantial investment involved and the three year delay in grape production while a new vineyard matures have, however, thwarted many farmers from converting to this crop. H~lrogeologic Description All of the fresh ground water stored beneath Southold originates from precipitation which falls within its boundaries. Approximately 50 percent of the precipitation (total yearly precipitation averages 44 inches) recharges under natural conditions. Fresh ground water exists only to a maximum of about Z00 feet, and is underlain by salt water. Water which falls on the hydrogeologic center of the North Fork (the ground-water divide) moves vertically downward in the ground-water system. North and south of the divide, an increasing horizontal component is evident until, at some point, ground water moves horizontally. Beyond that point, closer to the coastline, ground water moves horizontally and discharges to the various creeks and bays to the south, or to Long Island Sound to the north. A sense of the time involved in the movement of ground water through this hydrogeologic system is important. It is estimated 'that water recharging near the ground-water divide will remain in the aquifer for about 100 years. This residence time is shorter for water recharging closer to the shorelines. The flow of ground water through the unsaturated and saturated zone is dynamic in nature, although flow, velocities are quite slow. The rate of downward flow through the unsaturated zone is dependent on precipitation rates, and is on the order of 10 feet per year. Downward velocities in the saturated zone beneath the ground water divide are approximately 0.1 ft/day, with horizontal flows estimated as 0.Z to 0.4 ft/day at locations towards the shorelines (Baler and Robbins, 198Z). These low velocities result in contamination remaining within the aquifer for extended periods, while chemicals are carried through the aquifer until eventually ~flushingu to surrounding salty waters. Little mixing occurs within the aquifer system. Therefore, contamination remains in somewhat distinct plumes which travel through the aquifer, eventually (many years later) discharging from the fresh ground-water system. New waters are continuously introduced through recharge. Therefore, the future ground-water quality is dependent on the quality of recharge. If recharge waters are of a high quality over an extended period of time, the aquifer will eventually be restored. Under natural conditions the aqUifer underlying Southold was filled with pristine ground water. Over the years, various land uses within the Town resulted in contamination of recharge water. A major portion of the aquifer has gradually been filled with this contaminated water, and it is now observed at many of the Town's drinking water wells. The North Fork Water Supply Plan, formulated by Suffolk County, has recommended treatment for private and public water supplies which are adversely affected by pesticide or nitrate contamination. It is acknowledged that this is a "bandaid" type solution to the water supply problem. The long term solution involves controlling the multiplicity of sources of all contaminants, including nitrate, various pesticides, and the ever-growing list of organic chemicals which are utilized by industry, commerce, and private households. The work conducted under the Southold Demonstration Site portion of the New York State Fertilizer/Pesticide Project considers some important current sources of contamination. Under the funding agreement with EPA, this project was required to investigate sources of pesticide and fertilizer contamination. The scope was expanded somewhat to include other sources of nitrate associated with lawn fertilization (runoff from impervious surfaces, natural vegetation, and domestic sewage). Overall ground-water pollution source 3 control must address these and numerous other contaminant sources and constituents. Southold's Ground-Watew Problems Southold's ground water is particularly susceptible to contamination for a number of reasons. The Town's soils are highly permeable~ allowing recharge waters and associated contaminants to leach rapidly downward. The relatively low topography of the North Fork and resultant shallow depth to the water table minimize the time required for recharge to reach the ground water. The physical dimensions of the fresh ground-water reservoir are quite limited, making contamination problems particularly serious. Fresh water exists only in relatively thin lenses (maximum thickness of about ZOO ft) beneath the landforms of Southold (Baler and Robbins~ 198Z). Ground water is the only source of potable water to the approximately Z0~000 residents of Southold and the large number of summer visitors. In addition, agriculture~ which represents a major portion of the economy, relies on fresh ground water for irrigation purposes. Due to this sole reliance on ground water and the lack of alternative sources~ ground- water protection needs to be of utmost priority in Southold. In recent years, a number of problems have arisen with the area's ground-water quality. A brief summary of existing problems follows: Three (14%) of the Z1 Suffolk County test wells in Southold have average nitrate concentrations higher than 10 rog/l, the drinking water standard. Nine of these wells, or 43 percent, have had at least one sample exceeding 10 mg/l nitrate, indicating that the problem is not an isolated one. 897 (16%) of the 5,784 Southold wells tested by the Suffolk County Department of Health Services exceeded the 7 ppb drinking water guideline for aldicarb, a pesticide used on potato farms from 1975 to 1979, and another 763 (13%) had detectable concentrations below this guideline. Figure I-1 shows the distribution of aldlcarb conta~nination of ground water in Southold. Carbofuran, another agri- cultural pesticide, was found in 598 (30%) of the 1986 wells sampled in the Town of Southold, with 1Z3 {6%) of these wells exceeding the 15 ppb drinking water guideline. Trace organic compounds have been detected in 3 (15%) of the Town's Z0 standard observation wells. These wells are part of a network installed by the Suffolk County Department of Health Services for various monitoring purposes. Leachate from the Southold landfill has caused standards for selenium, manganese, and iron to be exceeded in water obtained from downgradient wells. Salt water intrusion and up- coning has become a problem in some shoreline areas due to excessive ground-water withdrawals. two foremost contaminants of Southold's ground water. The project has sought to identify means of reducing nitrate concentrations in ground water through evaluation of the effects of various land uses and management practices on nitrate concentrations in recharge water. Since the primary sources of nitrate leaching are fertilizers and septic systems, the effects of various agricultural and residential fertilization practices and a range of housing densities have been evaluated. Aldicarb differs from nitrates in that it is no longer used on Long Island. The question of concern is therefore not how to manage aldicarb applications, but how long it will take for ground-water concentrations to diminish to acceptable levels. By tracking the attenuation and movement of aldicarb from the soil surface to and through the ground water, estimates can be made of how long it will present a problem in Southold. A technique is presented for screening of pesticides prior to use and for identification of those most likely to leach to ground water. Such a procedure could be used in the pesticide registration procedure as a means of reducing possible new ground-water contamination by pesticides. Because this project focused on fertilizers and pesticides, ground-water contamination by sources not directly related to housing or agriculture have not been addressed (e.g. landfill, industrial and commercial sources). However, since potential organic chemical contamination is an important consideration in planning residential densities, a summary of existing data on organic chemicals found in Long Island ground water has been included. Objectives and Limitations of this Study The Southold Demonstration Project has focused on nitrate and aldicarb, the Water quantity issues such as overpumping and salt water intrusion are also ignored in this report, but will need to be considered in conjunction with the LEGEND l~--I ~ 1.4 ppi} ~ 1.4 3.5 PPB ~3.6 7.0 ~ ~--~7.O ' 14.0 PPB 14.0 - 112.0 PPB (Concentrations In Parts Per Billion) FIGURE [-[, D!STRIBUTION OF ALDICARB CONTAMINATION OF GROUND WATER IN SOUTHOLD, N.Y. recharge quality data presented here when land use decisions are made in Southold. All of the calculations presented in this report are site-specific~ and extrapolation of the specific simulation results to other areas may be misleading. The nitrate recharge values are influenced greatly by soil type~ weather patterns, and a number of land management factors which have been made as specific as possible to the Southold study sites. The aldicarb calculations were performed for a narrow transect across the North Fork and are dependent on the aldicarb application schedules, water table depths, and ground-water flow patterns for that particular site. This report provides information useful for land use planning, agricultural and turf management practices, public education efforts, and continued research aimed at restoring and protecting Southold's ground-water quality. Specific recommendations are presented in Chapter VIII. In addition, the work conducted in the Town of Southold and presented in this report demonstrates the computer modelling capabilities developed by Cornell University's Center for Environmental Research. These methodologies are transferable outside of the specific study area, provided that accurate site specific data are available for use in the model. Chapter li AN INTRODUCTION TO THE WATER AND LAND RESOURCE ANALYSIS SYSTEM The purpose of the Water and Land Resource Analysis System (WALRAS) is to make it possible to compile available data on water and dissolved chemicals and to perform the calculations necessary to evaluate existing and future sources of ground-water contamination. Use of WALRAS for simulation of water and contaminant movement to and through the ground water relies on site-specific data for climate, soil types, land uses, and management practices. The WAI~q~.S Process The WALRAS process consists of deriving a site-specific budget for water, taking into account precipitation, evaporation~ plant uptake, runoff, and leaching~ then using this water budget for calculation of movement of the contaminant of interest, in this case either nitrate or aldicarb. WALRAS simulations consist of calculation of water and contaminant movement through three zones: (1} the root zone, characterized by presence of plant roots, organic matter, and microorganisms, (Z) the unsaturated zone, with less organic matter and little or no microbial activity, and (3) the saturated zone, or ground-water layer (Figure H-l), where both biological and chemical activity are generally believed to be low. Movement of chemicals downwards from the root zone depends on how soluble they are in water~ how quickly they break down in soil, and how strongly they adsorb to soil particles. Both nitrate and aldicarb are highly soluble and only very weakly adsorbed to soil particles, so both travel freely from the root zone with recharge and irrigation water. Given the aerobic condition in the Long Island soil-water system, it is believed that nitrate concentrations diminish little if at all after the nitrogen leaves the root zone. Therefore~ WALRAS simulations of nitrate contamination can be confined to this one zone with the assumption that whatever nitrate leaves the root zone will eventually reach the water table (Figure n-Z). Within the root zone~ mass balance equations are used to represent the interactions between the various organic and inorganic forms of nitrogen in the soil and plant biomass. Inputs to the system include nitrogen contained in fertilizers~ precipitation, and animal wastes~ and outputs include gaseous losses, runoff, and leaching out of the root zone. Uptake and fixation of nitrogen by plants provide temporary sturage, with outputs when the plants die or are harvested. Unlike nitrate, aldicarb may degrade within the unsaturated and saturated zones, so all three zones need to be included in the simulations. As with nitrate, the first step is to create a water budget, in this case tracing water movement all the way to and through the aquifer. Next, the pesticide movement is simulated, based on water flow and how strongly the chemical is adsorbed to soil particles. Finally, degradation rates are calculated for aldicarb in each of the three zones. The highest breakdown of aldicarb occurs within the root zone~ near or at the soil surface. As microbial activity decreases with depth through the soil profile, degradation of aldicarb becomes slower. Since field and laboratory studies FIGURE ]]°1. ROOT ZONE, UNSATURATED AND SATURATED ZONES, COHTAHINAHT IPPLIC, ATION I ~ATER INFLUX. INCLUDING. ,~"~,,.~'~t RAIN, SNONHELT,, IRRIG~ AND OVERLAND FLON ROOT ZONE ~ ~.... UNSATURATED ZONE SATURATED ZONE EVAPOTRANSPIRATION LEACHING I I PERCOLATION LEACHING ~ PERCOLATION S~RFACE RUNOFF r Sandy Io~ to silty loam Significant Influenae by planta. have estimated a wide range of degradation rates, this is an area of uncertainty in our simulations and is represented by the range of results presented in Chapter VI. Interpretation of WALRAS Results Results from WALRAS simulations are usually summarized as annual averages, not upper limits for contaminant concentrations. An annual average of 10 mg/1 nitrate leaching to ground water means that the concentration leached would be expected to be higher than 10 my/1 about fifty percent of the time and lower than this value the other fifty percent. In order to prevent the 10 mg/1 drinking water standard from being exceeded this frequently, an average level substantially lower than the standard should be used for planning purposes. Statistical analysis of Nassau County well data has been used to compare mean nitrate concentrations of regions four square miles in size with the percentage of samples exceeding the 10 m~/1 drinking water standard (Porter, 198Z). This analysis showed that when the mean nitrate concentration was 6 my/l, the 10 my/1 standard was exceeded by ten percent of the samples. Similarly~ Table H-1 shows the relationship between annual average nitrate concentrations and the probability of individual samples not exceeding the 10 rog/1 standard. Table II-1 The Relationship Between Annual Average Nitrate Concentrations and the Probability that the Standard Will Not Be Violated by Individual Samples Probability of Not Exceeding 10 rog/1 Average Nitrate Concentration 50% 10 mg/1 90% 6 rog/1 99% 3 rog/1 99.9% Z rog/1 In planning future land uses for Southold, an explicit plamning standard for nitrate will need to be chosen which incorporates the margin of safety most acceptable to the Town. For example, if it is decided that ten percent of the time 9 the 10 mg/1 standard can be exceeded by individual samples, then a planning standard of 6 rog/1 would be used. Similarly, allowing samples to exceed 10 rog/1 only one percent of the time would require lowering the planning standard to 3 rog/1. Once the average concentration has been specified, the annual loading level compatible with the average concentration can easily be determined. This level can then be applied in reviewing individual sources of nitrogen. Many chemicals other than nitrate need to be considered in planning for protection of Southold's ground-water quality. Because of the lack of extensive well data for each of these chemicals, however, exact planning standards such as the one for nitrate cannot be derived. Protection of ground water from agricultural pesticides can be provided through thorough screening of proposed chemicals, as discussed in Chapter VII. The likelihood of ground-water contamination by organic chemicals is FIGURE 11-2, SOVEHENT OF NITRATE TO AND THROUGH THE GROUND HATER. ZONE related to housing density, but cannot be accurately quantified with existing data. The WALRAS simulations for nitrate can provide a base for planning decisions in Southold but should be supplemented with consideration of the potential for contamination by other chemicals as well. The New York State Department of Health drinking water standards and the New York State Department of lo Environmental Conservation Effluent Standards for discharges to ground water can serve as a starting point for development of criteria (see Table H-Z). As an interim planning guideline it would appear prudent to adopt loading levels for organic contaminants which correspond to less than 50 l~ercent of the maximum level specified for the contaminant in drinking water. Table II-Z New York State Standards for '#ater Quality for Contaminants Considered in This Report Contaminant NYS DOH NYS DEC Health Effluent Standard or Standard Guideline (if more stringent*) Nitrate 10 rog/1 -- Aldicarb 7 ug/1 0.35 ug/1 Benzene 5 ug/1 not detectable Carbofuran 15 ug/1 -- Carbon Tetrachloride 50 ug/1 5 ug/1 Chlorodibromomethane 50 ug/1 -- Chloroform 50 ug/1 -- Methylene Chloride 50 ug/1 -- Tetrachloroethylene 50 ug/1 -- Toluene 50 ug/1 -- 1,1,1 Trichloroethane 50 ug/1 -- Trichloroethglene 50 ug/1 10 ug/1 Xylene 50 ug/1 -- * The effluent standard is defined stringent standard is specified. to be the health standard unless a more The drinking water standards and guidelines apply to public water supply systems and were established to protect human health. The D.E.C. effluent standards apply to discharges to aquifers which are best suited to providing drinking water. Domestic on-site waste water disposal systems and agricultural practices are exempt from the effluent standards provided that these activities do not preclude the use of the ground water for drinking. WALRAS Applied to Southold In Southold, one of the major ground-water problems is nitrate contamination, and it is desirable to know 11 what the main sources of nitrogen are and how these sources can be managed in order to protect ground-water quality. WALRAS provides an organized means of computing the inputs and outputs of water and nitrogen from the land surface, with a focus on calculating the nitrate concentrations in ground-water recharge under a given set of environmental conditions and management practices. Another major ground-water problem in Southold is contamination by aldicarb, a pesticide widely used on potato fields from 1975 to 1979. Aldicarb use is now banned on Long Island, so the objective in this case is not management of its sources but determination of how far the contamination will spread and how long it will last. Two sites were chosen for WALRAS simulations in Southold. The first, bounded by Reeve's Avenue, Middle Road, Henry's Lane, and the Long Island Sou_nd, was used for the analysis of housing development presented in Chapter ITt. This site was chosen because it represents a typical area along Southold's shorelines where farming is currently the predominant land use but where pressure for housing development is now increasing. The second Southold site, a transect across the North Fork along Depot Lane, was used for estimation of nitrate leaching concentrations from a variety of agricultural crops as well as for evaluation of aldicarb movement and attenuation in the ground water (Chapters V and VI of this report). This site was chosen because it had recently been used for a field investigation by the Suffolk County Department of Health Services and therefore had detailed well data for nitrate and aldicarb that were unavailable at other locations. The data used in this study are described below. Climate Data Climate data for Southold were obtained from the weather station at the Greenport Power House. Average conditions were obtained by averaging simulation results for 1976, 1978, and 1980 since these years represent average, wet, and dry precipitation amounts and their average approximates the long term precipitation average at this station. For WALRAS calculations, each year is divided into 30 twelve-day timesteps to account for seasonal variations in climate. Soil Data In the Southold study areas, there are three major soil groups: Riverhead and Haven sandy loams, Carver and Plymouth sands, and Scio silt loam. The ltiverhead and Haven soils are well drained and moderately coarse textured. They make up the largest area of farm land in Suffolk County and are used extensively for growing potatoes and other vegetables. The Carver and Plymouth sands are excessively drained and coarse textured. Because fertility is low and permeability is rapid, these soils are not as well suited to agriculture but do make up portions of farm fields in Southold. Scio silt loam is moderately well drained and medium textured. It is found in small pockets among the Haven soils. Values used for the hydrologic properties of these soils are listed in Appendix B, Table B-1. Land Use Data In order to quantify the effects of existing and future land uses on ground- water quality at the Depot Lane study site, land uses were mapped using data from the Long Island Regional Planning Board, as well as extensive field checking. The 15 categories used are defined in Appendix B, Table B-Z. Residential land uses were further characterized according to their associated areas of turf, unfertilized vegetation, and impervious land surfaces for the purpose of compiling the land use effects on the water and nitrogen budgets. (These percentages are listed in Apendix B, Table B-3.) Land Management Data Management practices play a key role in determining water and chemical budgets in the root zone. The rate and timing of fertilizer applications, for example, determines how much nitrogen is added to the soil and whether it is present and available at a time when plants can use it. For each crop simulated, we estimated average fertilization practices by consulting with the Suffolk County Cooperative Extension Association, an agency that makes local fertilizer recommendations and which works closely with Southold farmers. These fertilization rates and schedules are shown in Appendix B, Table B-4. Typical planting and harvest dates for each crop appear in Appendix B, Table B- 5. Between planting and harvest, estimates must be made of the rate of plant growth as it relates to the ability of each crop to capture available soil nitrogen (Appendix B, Table B-6). Wherever possible, field data have been used to obtain these numbers. For crops where no data are available, estimates of nitrogen uptake rates have been made based on physical growth curves. 13 Chapter HI EFFECTS OF HOUSING DEIrELOPMENT ON GROUND-WATER QUALITY Housing development in Southold needs to be carefully planned for protection of ground-water quality. Residential septic systems and lawn fertilizers can be major sources of nitrate nitrogen, and household organic chemicals also can contaminate underlying ground water. Increased housing density will usually increase contamination. Other land uses associated with residential development will also add to the potential sources of ground-water pollution. Low housing density combined with restraint in lawn fertilization will limit the impacts associated with housing development. Lawn Fertilization Rates Suffolk County Cooperative Extension recommends fertilization of home lawns with either 1.0, Z.0, or 3.0 lb N/1000 sq. ft, depending on type of grass and the degree of greenness desired by the homeowner (Suffolk County Cooperative Extension, 1981). A survey conducted in 1980 by the Cernell University Center for Environmental Research found that 45 percent of Southold's home owners fertilize their lawns with an average of Z.4 lb N/1000 sq. ft./yr. and the remaining 55 percent use no lawn fertilizers at all (Pike et al.~ 1980). Using WALRAS, we simulated the nitrate leaching concentrations resulting from use of 0, 1, 2, or $ lb N/1000 sq. ft. on home lawns~ with 50 percent of the grass clippings removed (Table HI-l). In each case~ it was assumed that half of the fertilizer was in fast-release form~ and the remaining half was in a slow-release organic form providing nitrogen later in the growing season. As shown in Figure rrr-1, nitrate · leaching steadily increases with increasing doses of fertilizer nitrogen, and rates higher than 2 lb N/lO00 sq. ft. result in leachate concentrations -exceeding the 10 mg/1 drinking water standard in ground water immediately underlying the lawn. ~ s. S 0 I 2 3 FIGURE III-l. NITROGEN CONCENTRATIONS OF RECHARGE TO GROUND WATER FROM VARIOUS LAWN FERTILIZATION RATES, These results are confirmed by field experiments conducted at the Long Island Horticultural Research Laboratory (Snow~ 1976). A blend of four Kentucky Bluegrass varieties was planted in the spring of 1974 and fertilized at rates of 0~ 2~4~ and 8 lbs./1000 sq. ft. divided into 0, 1, 2, 4~ or 8 applications per year for the following two summers. During the second summer~ an average of Z5 percent of the applied nitrogen was leached from the root zone when the grass clippings Table HI-1. Rates and dates of lawn fertilization used in WALRAS simulations Total N Fertilization Applied Rate Per Per Year Application (lb N/1000 (lb N/1000 Dates of scA. ft.) Source scI. ft.) - Application 0.0 1980 Survey 0.0 N.A. of Town of Southold* 1.0 Suffolk 1.0 Sept. 10 County C.E. recommendation Z.0 Suffolk 1.0 May 13 County Sept. 10 C.E. recommendation 2.4 1980 Survey 1.Z May 1 of Town Sept. 10 of Southold* 3.0 Suffolk 1.0 May 13 County Sept. 10 C.E. Nov. Zl recommendation * A 1980 Survey of the Town of Southold showed 55% of the homeowners using no lawn fertilizers and the remaining 45% using an average of Z.4 lb N/1000 sq. ft./yr. were removed, and 50 percent when the clippings were left in place. As the turf became more mature, the leaching rates were expected to increase further still because the net amount of nitrogen incorporated into the plant biomass each summer would diminish as the grass reached an equilibrium in growth. It was concluded that when clippings are left in place, turf fertilization at rates higher than Z lb/1000 sq. ft./yr, is likely to result in ground-water recharge concentrations exceeding 10 rog/1. Removal of grass clippings reduces the amount of nitrate leaching from the lawn, but the clippings will still cause nitrate leaching in the landfill or wherever they are disposed. A better solution is to use them as mulch or compost to provide nitrogen to other plants and reduce the need for other fertilizers. Analysis of aerial photographs (Pike et al., 1980) has shown that the portion of each lot covered by turf varies according to the lot size. With one-acre or larger lots in the Town of Riverhead, for example, an average of only 16 percent of the area is covered by turf and 0Z percent is devoted to other types of vegetation, presumed in our simulations to be unfertilized. To whatever extent fertilizer is used on gardens, shrubs~ or other plantings other than lawns, our leaching rates in these simulations would be underestimated. With l/Z- to 1/4-acre lots, the turf increases to an average of 45 percent of the area and unfertilized vegetation drops to only 17 percent. Since water percolating through unfertilized areas leaches much less nitrogen than that draining turf, the effects of lawns on ground-water quality depend on the total area they cover as well as on lawn fertilization rates. The following section takes these varying amounts of turf into account in calculations of the nitrate leaching concentrations produced by various densities of residential development. Ho,.,,i,,~ Densities Southold currently is zoned at two acres per house. The Town allows the total number of permissible houses to be clustered into 50 percent of the land area. For example, a developer who has 100 acres of land could build 50 houses, either with two-acre lots spread evenly across the development, or with one-acre lots clustered onto half the land and the remaining 50 acres left as open space.* The Town Planning Board is considering allowing the houses to be clustered even closer together, into 35 percent of the total land area, for a net housing density of 1.4 houses/acre. 15 The nitrate leaching concentrations produced by various housing densities in Southold are shown in Figure rrr-z, with assumptions as outlined in Table HI-Z.** As housing density increases~ nitrate concentrations also increase~ largely because of the higher amounts of wastewater generated. The contributions from on-site wastewater disposal rise in proportion to the number of people living in the area, and make up a progressively larger portion of the total nitrogen leached as the density of housing increases. Inputs from lawn fertilization, the other major nitrogen source, also increase as residential density increases up to 3 houses/acre, after which the smaller lot sizes lead to less area in turf (Table I~-Z}. The amount of lawn fertilization has a large effect on the nitrate leaching concentrations, as shown in Figure I]I-3. At a housing density of one house per acre, for example, the concentration ranges from 2.7 to 5.4 rog/1 as lawn fertilization is increased from 0 to 3 lb N/1000 sq. ft. For the housing density calculations in this report, it was assumed that 45% of the lawns are fertilized with 2.4 lb N/1000 sq. ft. and that the remaining 55% are unfertilized. This is an approximation for Southold, based on a 1980 survey of 53 households in the town (Pike et al., 1980). Determination of the maximum density of housing below which ground- water quality in Southold would be preserved depends in part on the decision regarding safety margins. Table 1~-3 compares housing densities in Southold with the percentages of time that each would be expected to produce leaching concentrations higher than the 10 rog/1 *Actually, only 40 houses could be built under the clustering option because 20% of the land would need to be reserved for roads, curbing, parking, etc. **All housing density simulations were based on year-round rather than seasonal occupancy. standard. Determination of which of these levels are acceptable is a public policy decision that should be made by the Town before long-range land use plans are completed. The values represented in Figure I~-Z and Table rrr-3 are based on the specific set of assumptions listed in 16 Table III-Z. If variables such as the percentage of each lot maintained as turf or the average amount of fertilizer used on lawns were to change, then the WALRAS calculations would need to be revised accordingly. 0.5 1.0 ~ 2.0 2. 4.0 FIGURE III-2. THE EFFECT OF HOUSING DENSITY ON NITRATE CONCENTRATIONS LEACHING TO GROUND WATER IN SOUTHOLD, N,Y,~# *THESE DENSITIES ARE INCLUDED BECAUSE THEY ARE USED IN THE FOLLOWING CLUSTER VS. SPRAWL SECTION OF THIS REPORT, **IN THESE SIMULATIONS, 45[ OF THE LAWNS ARE FERTILIZED WITH 2.4 I~ N/lOeO SQ. FT., AND THE REMAINING 55[ ARE UNFERTILIZED. l? Table III-Z Assumptions related to nitrogen inputs for WALRAS simulations of residential development · 45% fertilized with 1.Z lb N/1000 sq. ft. on May 13 and another 1.2 lb N/1000 sq. ft. on September 10. · 55% receiving no fertilizer. · Lawn fertilizer divided into half fast-release form and half slow- release organic form for uptake later in the growing season. · Turf mowed every 12 days during the summer, with removal of 50 percent of the clippings. Res/dential Land Cover Fractions2 % Other Houses/Acre People/House1 % Turf Vegetation3 % Impervious 0.5 Z.54 18 61 1.0 Z. 54 25 50 1.4 2.54 32 39 29 Z. 0 Z. 54 38 28 34 Z.9 Z.54 45 17 38 4.0 2.54 40 14 46 1) From: 1980 Census for the Town of Southold. 2) From air photo analysis, unpublished data. 3) Vegetation other than turf is assumed to be unfertilized. In cases where gardens or other plantings receive fertilizer, our simulation results may underestimate nitrate leaching concentrations. 18 Leached Img/I | KE__Y FIGURE III-3. SIMULATED NITRATE LEACHING CONCENTRATIONS FROM A RANGE OF HOUSING DENSITIES AND LAWN FERTILIZATION RATES, Table rrr-3 Comparison of Southold housing densities with the percentage of time that the 10 rog/1 nitrate standard can be expected to be exceeded. With 45% of Lawns lle ceiving With All Z.4 lb N/1000 Lawns Housing With No sq. ft. Receiving Density Lawn and 55% 3 lb N/1000 (houses/acre) Fertilization Receiving None sc],. ft. 0.5 0-1% 0-1% 1-10% 1.0 0-1% 1-10% 1-10% 1.4' 1-10% 1-10% 10-50% Z.O 1-10% 1-10% 10-50% 2.9* 1-10% 10-50% 10-50% 4.0 10-50% 10-50% 10-50% * These densities are included because they represent the net housing densities created by clustering Z-acre or 1-acre zoning into 35% of the land area. Cluste~ vs. Sprawl Development A subarea of the Town of Southold was chosen for simulation of the effects of clustered housin~ development on nitrate concentrations in ground-water recharge. Bounded by Reeves Avenue, Middle Road, Henry's Lane, and the Long Island Sound, this subarea is portrayed in Figure III-4. Currently this region is zoned at one house per two acres, but Southold allows clustering into 50% of the land area, so a net housing density of one house per acre is allowed in the developed portion. Usually in a cluster development the remaining land would be left as open space covered by unmaintalned vegetation such as woodlands or meadows rather than lawns or crops. This provides recreational space as well as a buffer area to help compensate for the increased impact to ground water of spacing the houses more closely together. In Southold, one of the reasons for considering denser spacing of houses is to make it possible to preserve valuable agricultural land in inland areas by clustering houses along the shoreline. This does not fit the traditional concept 19 of clustered development, however, since the open space would be in another intensive use rather than left as vacant land. Figure r~-5 shows the results from simulations representing clustered housing development at densities of either one or two houses per acre. When the remaining half of the land area is left in unfertilized vegetation, the nitrate leaching concentrations remain low (Figure ]]]-5 a and b). If this land is used for agriculture instead, however, the concentrations increase two or three-fold to levels exceeding the 90% compliance level of 0 rog/1 (Figure rrr-5 c and d). Southold is considering changing its cluster ordinance to provide more open space by allowing closer spacing of houses, using 35% rather than 50% of the total land area. As shown in Figure IH-6, the nitrate leaching levels become higher still if 35% of the land area is used for clustered housing and the remaining 65% for agriculture. FIGURE [II-Z{; SUBAREA OF THE TOWN OF SOUTHOLD, LONG ISLAND, N,Y, 2O Nitrate Leaching 7.8 rng/I FIGURE 11I-5. THE EFFECT OF CLUSTERED DEVELOPMENT ON THE NITRATE CONCENTRATION OF RECHARGE WATER, HOUSES ARE CLUSTERED INTO 5~ OF THE LAND AREA, WITH EITHER AGRICULTURE OR UNFERTILIZED VEGETATION ON THE REMAINING ACREAGE, Filling the open space with agriculture is contrary to the usual definition of cluster development. The basic concept behind clustering is that land is left in open space to compensate for the increased housing density. In Southold, where nitrate and pesticide contamination of ground water are problems, if housing is allowed to be ~lustered then the remaining open space should be reserved for unfertilized vegetation or other uses which would provide the highest quality recharge water. !The high fertilizer and pesticide requirements of land uses such as golf courses or some types of agriculture make them intensive uses in & b ~ d P= potato fields H-housing V= vacant land (b) Net Housing Density- 1.4houses/acre {c} Net Housing D®nslty=2.0hOUSeS/aCre (d) Net Housing Denlity-2.gho~aes/acre FIGURE III-6, THE EFFECT OF VARIOUS CLUSTERING OPTIONS ON NITRATE CONCENTRATIONS LEACHING TO GROUND WATER FOR A SUBAREA OF SOUTHOLD terms of ground-water quality, so they should be associated with low density housing rather than high density clustered developments. Recommendations for converting land from other uses back to natural, unfertilized vegetation can be obtained from Cooperative Extension, the Soil Conservation Service, and the Nonpoint Source Handbook in preparation by the Long Island Regional Plannin§ ]~oard (LIRPB, 1983). Using WALRAS estimates of the amounts of water and nitrogen recharged from potato farms, unfertilized areas, and 21 residential developments of various densities, it is possible to calculate what combinations of these land uses would provide recharge that would meet whatever nitrate standard is set. The problem with this approach, however, is that it would provide only an average nitrate concentration for the entire study area shown in Figure lII-4, encompassing nearly 1500 acres and a variety of land uses. The homeowner whose well consistently exceeds the drinking water standard for nitrate will not be consoled by the fact that his neighbor's well is consistently low, providing an acceptable average for the area as a whole. A more rational approach to planning for ground-water quality is to decide that each individual land use must meet certain quality standards. Since ground-water flow is made up of water trickling slowly through small underground pores, it does not involve much mixing as in surface water rivers and streams. As shown in Figure Trr-7, the effects of different land uses remain distinct within the ground water, so the quality of water obtained at any given well depends on its placement and depth in relation to these ground-water flow paths. The recharge concentrations from various land uses should therefore not be averaged as if the recharge water from these sources were mixing in a vast underground reservoir. Instead, each land use should be analyzed individually to see whether quality standards can be met. Housing densities can be analyzed in this manner using calculations such as those represented in Figure HI-Z. The potato farming results in this chapter are based on an average fertilization rate of 175 lb N/Ac at planting. The effects of varying this rate and timing are discussed in the following chapter, and the nitrate concentrations produced by other land uses, such as orchards, vineyards, or horse pastures, are presented in Chapter V of this report. 22 FIGURE HI-7. EFFECT OF LAND USES ON GROUND-HATER QUALZTY. ._:. ,,-'-' - · . · : . ,',.~. Infuence of potato fred on saturated zone · l'(esidential developments can contaminate ground water with chemicals other than nitrate. Use of nitrate results should therefore be only an initial step in planning land uses that are compatible with ground-water quality. Consideration also needs to be given to providing protection against contamination by synthetic organic chemicals contained in household~ commercial, and industrial products. Potential Contamination by Organic ChemicaLs Organic chemicals contained in certain household products have been observed in ground water underlying residential areas on Long Island. These chemicals enter ground water through the on-site domestic wastewater disposal systems~ leaks from underground storage tanks~ and spills on the land surface. In order to assess the impact of these chemicals from residential areas on ground water~ the four Long Island communities of Mastic~ Mastic Beach~ Wading l~iver and Rocky Point where organic contamination has been observed were selected for study. (Figure I]I-8). These communities were selected because on-site domestic wastewater disposal systems are used throughout each~ and because there is little or no industry so the organic contamination in each must be due to the residential and commercial activities. Most residents of these communities rely on private house wells for their water supply. Much of the water drawn from wells in each community is water that has been recharged from land in the same community~ although to some degree, water recharged outside of each community also is used. ~ocky Point and Wading River are located on the north shore and do not extend very far inland. Hence some of the deeper wells in these communities are probably withdrawing water which originated as recharge further inland. Mastic and Mastic Beach are adjoining communities - Mastic Beach is located on the south shore and Mastic is located inland of Mastic Beach. Most of the wells in Mastic probably draw water which was recharged there, and wells in Mastic Beach obtain water which was recharged in either Mastic or Mastic Beach. The Suffolk County Department of Health Services began testing water from individual home wells for these chemicals in the late 1970's and since 1981 has tested for the presence of 15 organic contaminants (Benzene, To|uene~ m- xylene, p-xylene, o-xylen% chloroform, bromoform, bromodichloromethane, chlorodibromomethane, carbon tetrachlorid% methylene chloride, trichl0roethylen% tetrachloroethylen% trichlorotrifluoroethane, 1,1,1 - 'trichloroethane). In 1981 and 198Z the 23 County Health Department collected and tested approximately 1000 water samples from these four communities for organics. The assessment used in this study is based on the 1981 to 198Z data set. The ten chemicals listed on Table TTT-4 were detected in one or more water samples from these communities. All of these chemicals can produce immediate adverse effects in humans if ingested in large doses (National Research Council, 1977). These chemicals may also be carcinogenic in small doses. The doses which a person would receive from drinking ground water contaminated to the levels found during the Health Department's sampling program would not induce accute toxic effects but might cause cancer (National Research Council, 1977). Table III-5 shows the number of wells where each chemical was found. Twenty-four percent of the well water LONG ISLAND SOUND ATLANTIC OCEAN FIGURE III-8. FOUR LONG ISLAND COHHUNITIES TESTED FOR CONTANINATION OF GROUND NATER BY ORGANIC CHEHICALS, Table I~-4 Description of Organic Chemicals Found in Ground Water Underlying Residential Communities on Long Island~ New York Contaminant · Chloroform · Trichloroethylene · Tetrachioroethylene Possible Solvent · Xylene Gasoline · Methylene Chloride Source: National Research Council~ 1977. Micrograms per liter (ug/1) is the same as pa~ts per billion, Guideline for Maximum Concentration in Drinkin~ Water 50 ug/1 50 ug/l 50 ug/1 50 ug/1 50 ug/1 50 ug/1 50 ug/1 50 ug/1 50 ug/1 Suspected Adverse Health Effects* Observed to cause cancer (leukemia) Toxic at high doses Narcotic effects at high doses Insufficient data on long term effects Toxic at high doses Toxic at high doses Insufficient data on long term Suspected of causing cancer Toxic at high doses May cause birth defects Insufficient data on long term effects Insufficient data on long term effects 25 Table IH-5 Number of Wells Affected by Organic Chemical Contamination in Wading River, Rocky Point, Mastic and Mastic Beach Contaminant Number (and Percent) Number of Wells Highest of Wells Where Detected Concentration Where Detected Above Guideline Found Benzene 4 (.4%) 4 (.4%) 170 ug/1 Toluene 5 (.5%) I (.l%) 61 u~/1 Chloroform 14 (1.4%) 0 19 ug/1 Trichloroethylene 17 (1.7%) 1 (0.1%) 110 u~/1 Tetrachloroethylene 51 (5.1%) 4 (0.4%) 1100 ug/1 1,1~l-TricMoroethane 103 (10.3%) 13 (l.3%) 330 ug/1 Carbon Tetrachloride I (0.1%) 0 7 ug/1 O-Xylene 1 (0.1%) 0 8 ug/l Chlorodibromomethane 1 (0.1%) 0 ?- ug/1 Methylene Chloride 1 (0.1%) 1 (0.1%) 80 ug/1 samples contained detectable amounts of at least one organic chemical and 3.7% contained one or more chemical at concentrations greater than the State Department of Health guideline for the maximum concentration allowable in ch'inking water. The chemical found most often was 1,1~1 trichloroethane~ a major ingredient in cesspool cleaners. Xylene, chlorodibromomethane~ and methylene chloride were each found in one sample. Benzene~ a known carcinogen and the contaminant with the strictest guideline~ was found in 0.4% of the samples. The data set presented here is not large enough to make any general conclusions about the percentage of wells contaminated in excess of the guideline. It should be noted that drinking water wells usually give biased samples and are not a good guide to assessing overall ground-water conditions. Mastic Beach has a higher percentage of wells with contamination in excess of the guideline than the other communities~ and this may be related to several factors including the high density and the fact that there is no non-residential area adjoining Mastic Beach from which uncontaminated water could come. The four communities studied are of different average housing densities. Figure HI-9 graphs housing density versus both the percentage of wells tested where organics were found and the percentage of wells tested where organics were found at concentrations greater than the guideline for drinking water. Table IH-6 26 Organics Detected Above Guidelines Mastic · Rocky Point FIGURE III-9. HOUSING DENSITY COMPARED WITH ORGANIC CHEMICAL CONTAMINATION OF GROUND WATER IN FOUR LONG ISLAND COMMUNITIES, Table Ill-6 Comparison of Average Housing Density With Organic Chemical and Nitrogen Contamination of Ground Water in Four Long Island Communities Percentage of Wells Sampled Percentage of Where Organics Average (and number) Were Detected Nitrogen Average Number of of Wells at Concen- Number of (Nitrate pitts Housing Density Wells Sampled Sampled centrations in Wells Sampled Ammonia) presents the data on which the graphs were based. From the graph it appears that the percentage of wells affected in a community is directly proportional to the housing density. This observation supports the theory that each house with its own on-site sewage disposal system is a potential source of organic contamination. Table HI-6 also shows the average nitrogen concentration measured in each community over the same period. As is expected the average nitrate level increases with housing density. However the presence and concentrations of organic chemicals in individual samples are not correlated with nitrate concentrations in the samples. Many of the water samples contained high concentrations of organics and a low concentration of nitrate. It is stressed~ however~ that drinking water wells are not located according to a pattern which would best measure the impact of sources of nitrogen. In fact~ the intent of locating drinking water wells is to avoid contamination. 28 29 Chapter 1V EFFECTS OF POTATO FARMING ON GROUND-WATER OUALrrY The well drained soils in Southold have required fertilization ever since the native Indians used fish to provide nutrients to their crops of corn, beans, pumpkins, and squash. Early English settlers had little manure or other fertilizers available and frequently had to abandon pastures when the soil lost its natural fertility. Introduction in the early 1800's of commercial fishing as a source of fertilizer produced dramatic increases in crop yields and started the trend of regular fertilization of Long Island crops. Other fertilizer sources included ~uano from Peru, wood ashes delivered by a sloop, and horse manure brought by rail from New York City. Commercial fertilizer became available in 1844, and formulation of fertilizers to meet Long Island conditions began in 19ll. Invention of the mechanical potato digger in the 1880's provided an important boost to potato farming, and by the 1900's growth of potatoes had become a major crop on Long Island. Potato Fertilization Rates A 19Z9 survey of Long Island potato farms showed the average fertilization rate to be 100 lb N/A (pounds of nitrogen per acre), with over half the farmers also applying 7.7 tons of manure (Underwood, 1933). By 1975, potato fertilization rates had vastly increased, with most farmers applying from Z00 to 150 lb N/A (Chu and Selleck, 1977). Other nitrogen sources, including other agricultural crops, residential septic systems, and lawn fertilization, also had increased, and nitrate pollution had been identified as a major problem in Long Island ground water. Since then much research has been carried out with the aim of making potato fertilization rates compatible with ground-water quality. As a consequence, Cornell recommendations for fertilization of Long Island potato fields dropped from a range of 175 to 140 lb N/A before 1975 to a range of 125 to 175 lb N/A from that date on. Typical current practice for Southold potato farming is to apply an average of 175 lb N/A at the time of planting in April or May, with individual farms ranging from 150 to slightly over 100 lb N/A. In order to analyze the effects of potato fertilization on ground-water quality, we used the Water and Land Resource Analysis System (WALRAS) to simulate five potato fertilization rates under climate and soil conditions typical of Southold. These five rates were designed to represent a range encompassing average current practice as well as the extreme low and high ends of the scale, ranging from 100 to 300 lb N/A applied at planting. Two different soil types were used in these simulations. The first, Riverhead and Haven sandy loams, represents a group of soils which are prime for agriculture and which are found on most of Southold's farms. The second, Carver and Plymouth sands, represents a group of soils which are much less fertile and are often used for farming when they form portions of a field with a better soil type. These soils also are much more highly permeable than the Riverhead and Haven sandy loams and therefore much more susceptible to leaching of fertilizer nitrogen. As expected, our simulations show less nitrogen leaching when less fertilizer is applied (Figure IV-l). On the Riverhead and Haven soils, the current average 3O 100 150 175 200 300 Fel'tlllzer Applied I lb N/A/Yrl FIGURE IV-l, THE EFFECTS OF POTATO FERTILIZATION RATES ON NITRATE CONCENTRATIONS LEACHING TO GROUND WATER FOR TWO SOIL TYPES IN SOUTHOLD, N,Y, AVERAGE PRACTICE IN SOUTHOLD IS 175 LB N/A/YR ON RIVERHEAD AND HAVEN SANDY LOAMS fertilization rate of 175 lb N/A produces leaching concentrations around 8 m~/1, with a range of 6 to 11 rog/1 as fertilization is increased.from 150 to ZOO lb N/A. The concentration leached can be decreased to 4 mF>/1 if only 100 lb N/A are applied, but this fertilization rate is too low for satisfactory potato yields. The 300 lb N/A rate contaminates recharge with over 20 inF,/1 nitrate and obviously supplies far more nitrogen than the plants can use. On the Carver and Plymouth sands, leaching rates are dramatically higher and drop to 10 m~l only at the unacceptably low 100 lb N/A fertilization rate. Since these sandy soils do not produce a good potato crop, they are used only when they form portions of fields with a more fertile soil type. After an extensive study of nitrogen needs for optimal potato ~rowth consistent with ~round-water quality on Long Island, Meisinger (1976) recommended that Suffolk County growers reduce their fertilization rate to about 150 lb N/A. This was based on research on Katahdin potatoes, and the fertilizer needs of other varieties are slightly different. Rykbost et al. (1977) concluded, however, that the yields of Katahdin, Cascade, Superior and Hudson potatoes in Suffolk County were not significantly increased by raising fertilizer applications over 150 lb N/A. Figure IV-Z represents a compilation of over twenty years of research at the Long Island Horticultural Research Laboratory on fertilization of Katahdin potatoes, one of the most common varieties grown in Southold. Nitrogen was supplied in a variety of forms including urea, ammonium nitrate, ammonium sulfate, and mixed fertilizers, with both single and split applications. As can be seen in Figure IV-Z, the optimal fertilization rate under this range of 31 conditions was found to be around 150 lb N/A. Since potato yields do not increase beyond this point, the additional dollars spent on fertilizer detract from net profits for those fields. The optimal fertilization rate might be different for other potato varieties. Spit App~cafions Kossack and Selleck (1979) and other researchers have recommended that potato fertilization on Long Island be split so that a portion is applied at planting and the remainder in five or six weeks when the plants are growing rapidly and have a better chance of taking up the nitrogen before it is leached (Figure IV-3). Field research has verified that split fertilizer applications produce comparable potato yields to single applications, with reduced losses of nitrogen to the ground water (e.g. Rykbost et al. 1979). O- 100- 150 150- 175- 200- 250- 100 150 175 200 250 300 Fet'tilizsr Appli~l I lb N/A/Yr ) FIGURE IV-2. RESULTS OF FIELD EXPERIMENTS CARRIED OUT FROM 1956 TO 1978 AT THE L,I.H.R.L, ON YIELDS OF KATAHDIN. POTATOES AT VARIOUS FERTILIZATION RATES. 32 FIGURE IV-3. COMPARISON OF NITRATE LEACHING FROM HEAVY FERTILIZATION AT PLANTING VS. LIGHTER FERTILIZATION WITH SPMT APPLICATION PRECIPITATION PRECIPITATION Research on five common potato varieties at the Long Island Horticultural Research Laboratory (Baier~ 5.H., and K.A. Rykbost, 1976) has shown that an application of 50 lb N/A at planting is adequate [o carry the potato crop through the four to six inch plant height stage and that additional nitrogen needs to be applied prior to the period of most rapid uptake, approximately five to seven weeks after planting. Using WALRAS, we simulated four different schemes for applying 150 lb N/A on Southold potatoes: a) 150 lb N/A at planting b) 100 lb N/A at planting, and 50 lb N/A 5 weeks later c) 50 lb N/A at planting, and 100 lb N/A 5 weeks later 33 d) $0 lb N/A at planting, 50 lb N/A 5 weeks later, and 50 lb N/A 8 weeks after planting. The simulated split application schemes show substantial improvement over a single application in terms of both lowered leaching concentrations and increased tuber nitrogen contents (Figures IV-4 and IV-5). This is especially true for potatoes grown on Carver and Plymouth sands, where the highly permeable soil looses nitrogen rapidly unless the plants can assimilate it before it has a chance to leach. Timing of the nitrogen availability is critical for proper tuber development. Applying the fertilizer too late can stimulate top growth but retard tuber formation and development. Field studies at the Long Island Horticultural Research Laboratory have shown that the highest FIGURE IV-4. EFFECTS OF SPLIT APPLICATION OF FERTILIZER ON SIMULATED NITRATE CONCENTRATIONS LEACHING TO GROUND WATER FROM POTATO FIELDS IN SOUTHOLD, N,Y, #THESE REPRESENT SPLIT APPLICATIONS AS DESCRISED IN THE TEXT, 34 Fertilizer Applied I lb N/A/Yr ] FIGURE IV-5, EFFECTS OF SPLIT APPLICATIONS OF FERTILIZER ON SIMULATED TUBER NITROGEN CONTENT IN SOUTHOLDJ N,Y, *THESE REPRESENT SPLIT APPLICATIONS AS DESCRIBED IN THE TEXT, nltrogen_use efficiency for potatoes can be obtained by applying a portion of fertilizer at planting and the remainder from plant emergence to the six to eight inch stage. Effects of HeaT Spring It is not uncommon in Southold for heavy rains to occur in the Spring, leaving potato geowers wondering whether additional fertilizer applications are needed. In order to respond t# this question, we simulated potato growth for typical years with average and above average precipitation amounts, adding an extra 13 inches of rain in June. This extra June rainfall is similar to what happened in Southold in 198Z but is mere extreme than Southold's usual spring weather. Fertilization was set at 175 lb N/A at planting, an average figure for Southold. As shown in Figure IV-6, in an average year on Riverhead and Haven soils, approximately Z0 percent of the fertilizer nitrogen is lost to the ground water. The additional 13 inches of rain in June increases leaching losses to close to 30 percent of fertilizer nitrogen. In a typical wet year, fertilizer loss is 40 percent, and this increases to 48 percent with the additional June rain. In order to respond to the question of whether additional fertilizer is needed after a heavy spring rainfall, examination needs to be made of the effects on potato yields of the additional losses of fertilizer nitrogen. Table IV-1 shows the simulated tuber nitrogen s.nd leaching losses per acre averaged over three years of weather, with and without the extra June rain. The extra June rainfall makes a relatively small difference in the amount of nitrogen leached and the amount accumulated in the tubers. In all cases the reduction in tuber nitrogen is less than ten percent, with the highest loss occurring with the three-way split application because of the close proximity between fertilizer application and the heavy rain. Even with the extra rainfall, however, the split application schemes still are the best at keeping nitrogen in the root zone where it is available for uptake by plants. Ecmmmic Consequences FertiH~ation Practices of Various The economic consequences of various fertilization practices depend on the current prices of fertilizer, which are known at the beginning of the growing season, and on the market price for 35 potatoes that prevails after harvest. Growers must therefore make fertilizer decisions without knowing in advance what the economic consequences will be. In a study of four Suffolk County farms, Rykbost et al. (1977) compared usual grower practices with fertilization rates recommended by the Long Island Horticultural Research Laboratory. Fertilizer applications by the growers in the range of 185 to ZT0 lb N/A yielded one to five percent more of the largest category tuber (greater than 1 7/8 inches) than did fertilization in the lower recommended range of 133 to 154 lb N/A. Although these yield differences are too small to be statistically significant, they are useful as an example for comparing net profits. 80' 60 40' 20' FIGURE IV-6, RIVERHEAD May 1 EFFECTS OF HEAVY SPRING RAIN ON NITROGEN LOST FROM POTATO FIELDS GROWN ON RIVERHEAD AND HAVEN SOILS, SHADED AREAS SHOW ADDITIONAL FERTILIZER LOSS DUE TO THE EXTRA 13 INCHES OF RAINFALL IN JUNE. 1973 i wet year I June 1 I July 1 13 inche~ extra rein Aug1 36 Table IV-1. Simulated values for nitrogen leached and nitrogen content of potatoes grown on Riverhead and Haven soils. Results represent the average of three years of weather, with and without addition of extra spring rainfall. Nitrogen Leached (lb N/A) Nitrogen in Tubers (lb N/A) Fertilizer Without With an Extra Without With an Extra Application Extra June 13 Inches of Extra Jane 13 Inches of (lb N/A) Rain Rain in June Rain Rain in June 300 165 169 138 138 ZOO 78 83 130 129 175 61 68 125 1ZO 150 49 53 114 112 100-50 * 42 49 121 116 50-100' 38 45 1Z5 121 50-50-50* 35 43 1Z8 1Z1 *These represent split fertilizer applications, with a portion applied at planting and the remainder applied from 5 to 8 weeks later. Whether the additional yield of large potatoes is worth the extra fertilizer cost depends on the current market price for potatoes. It is evident in Table IV-2 that for 1976 fertilizer prices the grower at two of the four sites would have come out ahead by using less fertilizer, even at the highest tuber price analyzed ($4.00/cwt). At the middle and low tuber prices ($3.50 and $3.00/cwt), three out of the four growers would have had higher economic returns by spending less on fertilizer. Fertilizer and other prices have risen since 1976, and this past year Southold farmers lost money regardless of their fertilization rate because the $3.50/cwt price for potatoes was too low to meet their costs. In general, higher fertilization rates become economically more advantageous as tuber prices rise. Since market prices are not known until the end of the growing season and can fluctuate dramaticaily from one year to the next, however, growers must make fertilizer decisions without the benefit of this knowledge. The favorable yields shown at 150 to 175 lb N/A and the hazards to ground water at higher fertilization rates make this seem to be a good average to aim for for potato fertilization in Southold. Split application can further reduce the leaching losses, keeping nitrogen in the root zone where the plants can use it. Potato Pest Management In considering pest management as it relates to ground-water quality on the North Fork, there are several key factors that have to be taken into account. These include: the extent of farming in the area, the intensity of crop production, the production of potatoes as a monoculture~ and the nature of the sandy soils upon which the crops are grown. On the North Fork it would be difficult to overstate the importance of farming to the local economy. The fraction of Southold's land devoted to farming and the value of the crops produced, have always been high. For example, out of a total area in Southold of 35,600 acres, 30,400 acres were farmlands in the early 1900's {Bond, 1947). By 1940, the farmed acreage had decreased to just over 15,000, a fraction which was still close to 50% of the total. During the intervening years, the area farmed has slowly further decreased as the population of Southold has increased. Nevertheless, nearly 40% of the total land remains in farming. Since the 1930's, potato farming has accounted for about half the cropland as can be seen from Figure IV-7. For decades, the practice has been to grow potatoes continuously with little rotation. 37 This pattern of production takes advantage of the long mild growing season and the fertile soils found on the North Fork. The specialization of farmers in potato growing has developed a high level of proficiency, reflected in the high yields of Long Island potato farms. As shown in Table IV-3, the yields of potatoes on Long Island are consistently greater than the national average. Potatoes are grown on Long Island generally as a late summer or fall crop. Compared to similar areas elsewhere in the United States which grow late potatoes~ the average yield on Long Island was 157 cwt/acre compared to ZZZ cwt/acre for all late potatoes over the twenty-five year period from 1955. Both yields can be compared to the overall average yield for all potatoes grown in the United States of 7.15 cwt/acre. During the 7.5 year period~ Long Island was only below the average for the United States for six years. Considering the vagaries of the weather and pests~ this is a remarkable achievement. Table IV-Z. Economic consideration of p.tato fertilization rates. (Taken from Rybost et al,, 1979) $3.00/cwt $3.50/cwt $4.00/cwt Fertilizer Tuber Tuber Tuber Site (lb N/Ac) Cost~ $/A S/A* $/A $/A $/A $/A Balance Fertilizer cost (provided by Al]way at Long Island on 11/11/76 on bulk price). 8-16-8 $137.25/ton~ 10-Z0-10 $166.S0/ton~ 3- 18-9 $1i0.00/ton, NH4NO3 $153.00/ton~ Urea $170.00/ton. Based on ~umbo and A tubers sampled from 6 X 15-foot row. 38 20,000 15,000, lO, O0O. 5,000. 1940 195o 1960 1970 198o YEARS 20000 18000 16000 14000 12000 Z FIGURE IV-Z, CHANGES IN POPULATION AND ACREAGES OF FARM LAND AND LAND USED FROM POTATO PRODUCTION IN THE TOWN OF 'SOUTHOLD, N,Y, FROM 1930 TO 1980, Five of these six low years occurred in the first half of the 1970's, a period during which potato pests on Long Island had become particularly difficult to control. The monoculture potato farming on Long Island has fostered vigorous populations of various pests including the Colorado potato beetle, aphids, nematodes, weeds, and the potato late blight. The most voracious and potentially damaging of these is the Colorado potato beetle. Infestations of the beetle were especially damaging in the 1970's until the introduction of the pesticide aldicarb in 1975. In the five years up to 1976, Long Island potato growers achieved an average yield .of ZZ6 cwt/acre. During the next four years when aldicarb was used the average was very nearly 300 cwts/acre. Aldicarb was withdrawn from use on Long Island when its presence in ground water was confirmed in 1979. Another pesticide used against the Colorado potato beetle, carbofuran, also was withdrawn following the widespread detection of the pesticide in ground water. The fate of pesticides in the soil depends on their chemistry and the soil properties. Principal factors which determine the characteristics of the soil ecosystem are pH, soil water, organic matter and soil physical characteristics. On Long Island generally, and on the North Fork in particular, the soils have a low pH and the amount of drainage is high, with an average recharge of over Z0 inches per year. The soil organic matter is low, and the soils are sandy with little 39 Table IV-3 Per Acre Yield of Potatoes on L.L and in U.S. 1955-1979 Yield/acre L.L (Fall) U.S. (Fall) Total U.S. Year (1) (Z) (3) 1955 215 169 161 1956 240 191 176 1957 235 185 173 1958 250 196 181 1959 ZZ0 188 184 1960 270 185 184 1961 258 196 196 196Z Z85 197 194 1963 265 206 ZO1 1964 Z55 185 185 1965 290 214 210 1966 Z55 213 Z10 1967 Z35 Z1Z Z10 1968 265 Z14 214 1969 Z60 223 1970 Z70 Z33 ZZ9 1971 Z30 Z36 230 197Z 207 246 236 1973 Z15 Z33 Z30 1974 250 253 Z46 1975 Z60 264 253 1976 310 Z69 260 1977 315 Z70 Z61 1978 Z65 280 Z66 1979 295 Z77 Z69 Average Z57 ZZZ Z15 Standard deviation Z8.5 33.1 3Z.0 Source: U.S. Department of Agriculture. Potatoes and Sweet Potatoes. Crop Reporting Board, Washington, D.C. Long Island Fall Crop. U.S. Fall Crop including Long Island. All U.S. production. 4O clay. Farm soils contain little organic litter, and biological activity is generally low. In consequence, soluble pesticides whose decomposition depends primarily on biological activity can be mobile and persistent in Long Island soils. As a result, an increasing number of pesticides have now been found in ground water underlying or adjacent to farming areas. To the potato growers on the North Fork, the discovery of pesticides in ground water and the subsequent unavailability of effective pesticides, have been unforttmate. The farmers' attempts to control pests using less effective pesticides has inevitably increased the number and volume of sprays. For example~ surveys of farm costs on Long Island showed that the average cost of pesticides was $126 per acre in 1970, the first year aldicarb was widely used. In 1981 when aldicarb was no longer available the average cost was $Z91 (Synder, 1977 and 1982), an increase of over 130% in only five years. Inflation accounts for approximately 40%, and the remainder represents cost increases due to the change in pesticides and the increasing amounts needed to compensate for their lower efficacy. There is now a critical need to develop effective techniques to control pests if potato farmers are to sustain adequate production and remain in business. At the same time~ further serious contamination of ground water used for drinking water by new application of pesticides is unacceptable to all residents of the North Fork. Given this double constraint, the need to control pests, and the need to protect groUnd water quality, Cornell University and its Long Island Horticultural Research Laboratory have instituted the Integrated Pest Management (IPM) program, a program of research to discover new methods of pest controls for Long Island's potato farmers. These pest controls include biological, cultural, and genetic methods as well as chemical ones in order to provide control techniques with minimal effects on nontarget organisms and on the environment. Two overall types of methods for managing potato pests can be distinguished: (1) direct pest control techniques for use in potato production, and (Z) alternate cropping systems. Direct techniques include eliminating unnecessary chemical spraying by determining action thresholds and by conducting field scouting to determine precisely when these damage thresholds have been exceeded. Research is being carried out on biological control possibilities, including suppressing beetles with the Beauvaria bassiana fungus. Cultural practices such as killing the potato vines earlier in the season to reduce food available to the beetles also are being investigated. Genetic research is looking into ways of making the potato plant more resistant to beetles, including breeding for the sticky hairs present in some wild species which can trap insects and cause them to starve. Alternative cropping systems that could reduce the need for pesticides include crop diversification and annual crop rotations. Wright et al. (1983) report that on three farms out of four~ potatoes grown on fields previously used for another crop had lower populations of the Colorado potato beetIe than did fields on which no rotation took place. On the fourth farm~ the overall number of the Colorado potato beetle were so Iow that differences between the rotated and nonrotated fields were immaterial. Crop rotation therefore could reduce the need for pesticides by reducing insect numbers and by slowing the insects' development of resistance which necessitates increasing chemical doses over time. Crops to be rotated would need to be carefully chosen to maintain farm income levels, as addressed by Lazarus and White (1983). Further information on Long Island's IPM programs is contained in Appendix A. Chapter V EFFECTS OF OTHER LAND USES ON GROUND-WATER OUALITY The previous two chapters have addressed the effects of housing and potato farming on ground-water quality in Southold. Although these are the Town's two most prevalent land uses, many other uses exist now or are being considered for the future. In this chapter we present the results of WALRAS simulations of nitrate leaching concentrations from these other land uses. The numbers in this chapter represent broad estimates with much greater uncertainty than in the potato simulations because similar field data are ~not available for verification of our models. The results presented here therefore represent only a general indication of nitrate levels, and are meant to be a first step indicating where more work is needed. Exi~ Land Uses In order to simulate an example of existing land uses in Southold, we chose an area extending across the North Fork along Depot Lane, bounded by Alvah's Lane, Linden Avenue, and Cox Lane. Land uses in this area include a variety of agricultural crops as well as housing, a golf course, and horse pasture (Figure V- I). The soils are predominantly Riverhead and Haven sandy loams, with smaller areas of Carver and Plymouth sands or Scio silt loam (Figure V-Z). Simulated nitrate concentrations in recharge from these existing land uses are shown in Figure V-$. A large portion of the land area has recharge concentrations in the 7.5 to 10.0 rog/1 range. This area consists primarily of potato fields, and nitrate levels can be reduced by changing the fertilization practices, as discussed in Chapter IV. The areas with recharge concentrations exceeding 10 rog/1 are either located on Carver and Plymouth sands or are used for sod farms, nurseries, horse pastures, or fields of mixed vegetables or cole crops. Those with low nitrate recharge concentrations (less than 5.0 rog/l) are either vacant land, fields being rested by year--round growth of grain, or low-density residential areas. Specific recharge concentrations for each of these land uses are shown in Table V-1. It should be kept in mind, however~ that these numbers are somewhat speculative since for uses other than potato farming no field data are available for verification of the simulation results. Field data collected in 1981 by the Suffolk County Department of Health Services on nitrate concentrations in ground water below Depot Lane are shown in Figure V-4. As with the simulated recharge values, the ground-water data showed a wide spread of values, ranging from 0.3 to 13.0 rog/1 nitrate. The lowest of these occurred at the ground-water' divide, below an area used for vacant land and houses. The highest was below an area used for potato farming. Possible Future Land Uses Possible future land uses for the Depot Lane area include continuation of existing uses, changes in agricultural crops, or increases in residential development. Continuation of the existing potato farming, with revised fertilization practices as suggested in Chapter IV, would be consistent with ground-water quality. Only the portions of fields on Carver and Plymouth sands would still have unacceptably high nitrate leaching concentrations. Other agricultural crops such as mixed R2 Ap LEGEND A - AGRICULTURE Ap - Potato Av - Mixed Vegetables Ag - Grapes Ao - Orchard An - Nursery As - Sod Ah - Horse Pasture R - RESIDENTIAL R1 - O to 1 D.U./Acre R2 - 2 to 4 D.U./Acre Ap Ap Rd -M~ddle Rd Ap Ap Ap Ap Ap Ag Ag C - COMMERCIAL/INDUSTRIAL - INSTITUTIONAL Pk - PARK/GOLF COURSE V - VACANT RB - RECHARGE BASIN Ao AY Ap Ap V New Suftolk Rd FIGURE V-L. LAND USES IN THE DEPOT L~.NE TRANSECT AREA, SOUTHOLD, N.Y. HR HR HR LEGEND HR -RIVERHEAD and HAVEN SOILS PC -CARVER and PLYMOUTH SANDS SD -SCIO SILT LOAM O -OTHER (Beaches, Cut and Fill, Made Land, Tidal Marshes) HR PC HR ~ HR HR PC FIGURE V-2, SOIL TYPES IN THE DEPOT LANE TRANSECT AREA, SOUTHOLD, I~,Y, LEGEND ~ 0.0 - ~2.5 - ~5.0 - ~7.5 - 10.0 MG/L > 10.0 MG/L I--~ Not Simulated (Concentrations in Milligrams Per Liter) 2,6 MG/L 5.0 MGIL 7.5 MG/L FIGURE V-3, RFSULTS OF NZTRATE SZMULATIONS IN THE DEPOT LANE TRANSECT AREA, SOUTHOLD, N,Y, 45 Table V-1 Simulated nitrate leaching concentrations for various land uses in Southold, N.Y. Nitrate Leached (rog/l) Total Fertilizer Riverhead and Carver and Applied Haven Sandy Plymouth ( lb N/A/yr) Loam Sands Cro~)s Cole Cropsa 140 10. Z 1Z. 0 Mixed Vegetablesb 1Z0 10.8 1Z.0 Potatoes 175 8.1 16.1 Gralnc 0 Z.5 1.3 Nurseries Z50 11.5 14.Z Orchards 100 6.4 1Z.0 Vineyards 30 5.6 5.4 Sod 1st Year Znd Year Other Land Uses Z14 8.5 11.4 Z89 19.6 Z1.5 Horse Pasturesd 0 Z. 1 Z. Z Golf Coursese 63 7.6 8.5 Parks 48 4.1 4.3 Vacant Land (Unfertilized vegetation) 0 0.9 0.9 a) ~) c) d)' e) Cole crops are those in the cabbage family, including cauliflower, These numbers are averaged over the entire land area, but for simulation purpose~ they were broken into different fertilization rates for different 46 40 20 MSL -2oI -120 -140, -160 M&IN ROAD 7,5 0.3 Concentrations is ppm ~ { ~ ~ fo 1~ i2 f3 1~ i5 16 1:7 18 f9 20 DISTANCE FROM NORTH SHORE (1000 FT) North Fork vegetables or cole crops appear to produce recharge concentrations in the same range as those from potato fields, and vineyards and orchards appear a bit lower (Table V-l). Simulations of sod farms show extremely high leaching rates as a result of the large amounts of fertilizer applied for rapid growth of quality turf. Although residential development appears beneficial from its relatively low nitrate concentrations, housing and associated services such as gas stations, dry cleaning operations, or other commercial development can contribute substantial amounts of other chemicals to ground water, including organic contaminants as discussed in Chapter III. For protection of ground-water quality, residential development should therefore be kept at a low density, preferably in the range of one or fewer houses per acre. Because the highly permeable Carver and Plymouth sands are not highly fertile, they commonly are used for housing rather than agriculture when they cover a large enough area, as in the Cutchogue Harbor section of the Depot Lane transect (Figures V-1 and V-Z). The high fertilization and irrigation rates needed for maintaining lawns on these sandy soils probably leads to nitrate leaching concentrations far in excess of those predicted here and in Chapter IH for residential development, but the lack of field data on either turf or septic system functioning on the sandy soils precludes further refinement of our simulation results. In general, the sandy soils would best be kept in unfertilized open space uses such as vacant land or parks for protection of ground-water quality. The more fertile and less highly permeable Riverhead and Haven soils, on the other hand, are suitable for a wide range of agricultural and residential uses. 47 Chapter VI ALDICARB CONTAMINATION OF SOUTHOLD'S GROUND WATER Aldicarb Use in Southold Aldicarb is a highly toxic insecticide, marketed under the brand name of Temik, and used throughout the United States on crops including potatoes, cotton, sugar beets, and oranges. Potato farmers in Southold began using aldicarb in 1975 and found it highly effective in controlling the Colorado potato beetle and golden nematode. As can be seen from Table IV-4, there was a dramatic increase in yields during the four years aldicarb was used compared to the preceeding five years. However, aldicarb and its byproducts are highly soluble and weakly adsorbed to soil, and it was expected that they might not fully degrade before reaching ground water (Porter and Beyer, 1977). When this proved to be the case, aldicarb was removed from use on Long Island in 1979, leaving the potato farmers with no equally effective means of combating the beetle and nematode problems. Since that time, aldicarb contamination of Long Island's ground water has been found to be widespread. Of the nearly 14,000 Suffolk County wells tested as of Jane 1983, Z8.9 percent had detectable aldicarb concentrations with 15.1 percent exceeding the 7 ppb drinking water guideline for New York State; 15.5 percent of the 5,784 wells tested in the Town of Southold exceeded 7 ppb and an additional 13.Z percent had detectable aldicarb concentrations lower than this amount. In this project the goal was to trace the attenuation and movement of aldicarb to the ground water. Within the root zone and unsaturated zone, water movement was calculated first, then aldicarb movement was computed based on its solubility and adsorption coefficient. Depot Lane Transect In 1980 the Suffolk County Department of Health Services had a series of geologic test holes and observation wells drilled along Depot Lane in Cutchogue in order to obtain information about how aldicarb and other contaminants move through the North Fork's aquifer. The transect cuts across the North Fork from north to south and runs parallel to ground-water flow in both the northerly and southerly directions. Private wells were sampled by the County in the Spring of 1980, and the observation wells were tested a year later. The results are shown in Figure VI-1. Aldicarb contamination was found up to 100 feet below the water table near the center of the transect and at shallower depths (up to 40 feet below the water table) at the northern and southern extremities, with concentrations ran~ing from one to 149 ppb. Aldicarb Simulation The Depot Lane transect was divided into 13 areas (Figures VI-Z and VI- 3), and the farmers were stzrveyed about their aldicarb use on fields within each of these regions. Results are shown in Table VI-1. For our simulations, the adsorption partition coefficient was set at 0.09 in the root zone and at 0.0 below the root zoner indicating that only a small percent of the aldicarb contained in the soft water would be adsorbed to soil particles in the root zone, and none below this zone. Oregon Rd. Middle IRR Main ~ Road Road ,~ ,~ CUTCHOGUE L.L~ (4) (891 [551 (* IC *) (1) (,) {.) ~% CREEK~ NECK ~ {'1 40 84(*) (a 18 ~ ,{31) (83)(*~ X * / / UO 6O 40 2O -20 DISTANCE FROM NORTH SHORE (1000 FT) Figure VI-1. ALDICARB CONCENTRATIONS MEASURED IN WELLS ALONG THE DEPOT lANE TRANSECT, SOUTHOI_D, N,Y, (Taken from: ~e~ort on the Occurrence and Movement of A~ricultural Chemicals in Groundwater: of Suffolk County, Suffolk County Department of Health Services, August, 1982.) North Fork / / / / 2 4 8 II / ! ! FIGURE VI-2, AREAS USED FOR ALDICARB ANALYSIS ALONG THE DEPOT LANE TRANSECT, SOUTHOLD, N,Y, 5O DISTANCE FROM NORTH SHORE (1000 FT.) FIGURE VI-3. SIMULATED ALOICAR~ pLU~ES ALONG THE DEPOT LANE TRANSECT, SOUTHOLO, N.Y, Table VI-1 Aldicarb use on fields along the Depot Lane transect, 1975 to 1979 Area Aldicarb Applied (lb active ingredient/A) 1 0.5 0.0 0.0 0.0 0.0 0.5 Z 2.5 0.6 1.0 1.0 1.0 6.1 3 0.4 0.7 1.1 1.1 1.1 4.4 4 0.1 1.4 Z.3 2.3 2.3 8.4 5 1.1 1.9 Z.4 2.4 Z.4 10.2 6 0.5 1.8 Z.S 3.7 3.7 12.2 7 0.0 0.0 0.0 0.0 0.0 0.0 8 1.7 2.0 Z.O 2.9 2.9 ll.S 9 1.6 3.7 3.7 4.6 4.6 18.Z 10 2.1 2.6 Z.6 3.8 3.8 14.9 11 2.1 4.0 4.0 5.2 3.? 19.0 1Z 0.0 O.Z 0.2 O.Z 0.0 0.6 13 1.? 1.7 1.7 1.? 1.7 8.5 1975 1976 1977 1978 1979 Total These numbers were calculated from a linear regression relating adsorption coefficients to soil organic matter derived from the results of four field and laboratory experiments (Kain and Steenhuls, 1982). The greatest uncertainty in modelling aldicarb is its rate of attenuation in the root zone, unsaturated zone, and saturated zone. The half-life, or time that it takes for half of the aldicarb to decay to nontoxic byproducts, is shortest in the root zone where microbial populations speed the decaying process, and longest in the saturated zone because of the lower temperatures and low rates of microbial activity. Unfortunately, laboratory and field experiments on these half lives are few, and the results are inconclusive. A previous review summarizing studies on aldicarb degradation in a variety of soils showed ranges in half-lives from one to Z31 days (Kain and Steenhuis, 1981). In our simulations, a 50 day half-life was used for the root zone. This was increased to 100 days for the next 30 centimeters in depth because microbial populations diminish with increasing depth (Pacenka and Porter, 1981). For the remainder of the unsaturated zone, a half-life of 10 years was used. The 50 day root zone half-life is from Kain and Steenhuis (198Z). The other numbers were derived by calibrating our model with field data from a potato field on the North Fork studied by Intera (1980) and a deep soil core taken at the Long Island Horticultural Research Laboratory (unpublished da[a). The simulation results indicating the dates and concentrations of aldicarb leaching to the ground water are shown in Table VI-Z. According to our simulations, all aldicarb that will reach ground wate~ is estimated to do so by 1984. The average concentrations of aldicarb recharging ground water were derived for each area by calculating total amounts of aldicarb applied, subtracting the amounts taken up by plants or degraded in the root zone or unsaturated zone, then dividing.by the total water recharged. In all areas except 4~lZ, where very little aldicarb was applied, the simulations show average aldicarb concentrations reaching ground water to be well above the 7 ppb drinking water guideline. How long they will remain higher than the guideline depends on ground-water flow and on aldicarb's half-life within the saturated zone. For areas along the shoreline, the contaminated ground water will probably be nflushed out", or discharged to surface water in a shorter time than it would take for the aldicarb degradation to occur. For inland areas, aldicarb's half-life in ground water will determine how long the contamination will remain. Laboratory estimates of the half- life of aldicarb in ground water have ranged from less than one to greater than ten years, depending on pH, temperature, and experimental procedure (Chapman and Cole, 198Z; Hansen and Spiegel, 198Z and 1983; Lemley, 1983).* Table VI-3 summarizes the available laboratory data on half-Iff es of aldicarb and its degradation products in water at various pH's and temperatures. Field testing has shown that virtually all aldicarb oxidizes to the sulfoxide or sulfone forms before reaching ground water, so it is the half- lifes of these two compounds that are of concern in the saturated zone. By extrapolating the available laboratory data to 15oc and pH 6 to represent ground-water conditions typical Prior to the availability of this laboratory data, Intera (1980) had derived a zero decomposition rate, or infinite half-life for aldicarb, based on their simulation results rather than on any laboratory measurements. 52 Table VI-Z Simulation results showing dates that aldicarb will reach ground water and its average concentrations in ground water below the Depot Lane transect in Southold, N.Y. Year Year All Average Aldicarb of Aldicarb Aldicarb First Reaches Reaches Concentration Water Tablea Water Table ]Recharging Ground Waterb 1 1979 1979 41 Z 1979 1984 5Z 3 1979 1984 Z4 4 1979 1984 39 5 1979 1984 56 6 1977 198Z 6Z 7 N.A.C N.A.C N.A.C 8 1976 1981 78 9 1976 1981 110 10 1976 1981 99 11 1976 1981 118 1Z 1977 1980 5 13 1976 1981 6Z a) b) This gives the year in which aldicarb would first have reached the water table according to the simulation. Note that the greater the distance between the land surface and the water table, the longer the time between when aldicarb is applied at the surface and when it reaches the water table. This is the average concentration of aldicarb in water entering the saturated zone between the time when aldicarb first reaches the water table and the time when it has all reached the water table. c) Not applicable because no aldicarb was applied to this section. of Long Island~ estimates were made of the half-lifes of these two compounds. For aldicarb sulfoxide, a range of eleven to sixteen years was estimated, and four to nine years for aldicarb sulfone. These two compounds are found in approximately equal proportions in the ground water, so averaging their half-lifes yields a rough overall estimate of the breakdown rate to nontoxic compounds. In this manner an overall average half-life of ten years was obtained. A great deal of uncertainty is contained in this number both because of the numerous extrapolations required and because the laboratory experiments greatly oversimplify the processes at work under actual ground-water conditions. A ten year half-life would mean a period of decades before the aldicarb plume would decay below the 7 ppb drinking water guideline in ground water 53 Table VI-3. Laboratory data on half-lifes (in days) of aldicarb and its degradation products in water at various pYI's and temperatures. Source 5.5 6.0 7.0 7.5 8 8.5 Aldicarb 15°C 3Z40 1900 170 b ZS°C - 266 Z45 - 266 - a Aldicarb Sulfoxide 15oc 440 360 10 b - NA NA 73 - c - 4ZZ1 4ZZ 4Z d Z5°C - 679 161 23 - a - NA 154 ZZ - c Aldicarb Sulfone Sources: a) b) c) d) 15°C 450 - 125 5 b - NA Z7Z 5Z c - 1458 145 15 - d Z5oc 4Z0 77 10 a NA 84 10 c 597 60 6 d Chapman and Cole (1982) Hansen and Spiegel (1983) Hansen et al. (1983) (Note: Values labelled "NA" are not yet available but will be at the conclusion of the experiments.) Lemley (1983) (Note: These data are extrapolated from experiments conducted at pH's 11 to 13.) underlying the Depot Lane transect and in other areas of Southold which had similar aldicarb application schedules. This half- life estimate is too uncertain, however, for it to be reliably used for water quality planning purposes. As a result, Cornell University is embarking on a more extensive study which will help to better define how long aldicarb will remain a problem in Long Island's ground water. 54 55 Cl~pter VII PI~STI~E SC]~]~NING TO PREVENT GROOND-WAT~ CONTAMINATION Southold's ground water has become contaminated by aldicarb and several other pesticides, including carbofuran, dacthal, dinoseb, and 1,Z dichloropropane. For aldicarb alone, millions of dollars have been spent by government and industry to test, treat, and study the ground-water contamination. Ultimately, the cost of clean-up entails expensive treatment of drinking water supplies, a measure which in itself does little to discourage further degradation of ground- water quality. A far more cost-effective measure would be to prevent contamination through methods incl~img thorough screening of new pesticides before they are used. Traditionally, pesticide manufac- turers have used field experiments to evaluate the environmental safety of their chemicals. There are problems in relying exclusively on field work for individual locations and soil types to provide universally acceptable assessments of chemical fates. In particular, the spatial heterogeneity of physical, chemical and biological field processes, and the temporal variability of climate, make it difficult to obtain an accurate field description of pesticide distribution. It is virtually impossible to reproduce these variable field conditions in the laboratory, or to extrapolate laboratory results and apply them to field conditions. In addition, testing for contaminants in concentration ranges of parts per billion requires sophisticated technology, and in some cases the water quality standard for a chemical is lower than the analytical detection limits. There is a marked tendency to address these problems by doing more sampling, extending field experiments over longer periods, -and in general imposing greater requirements on pesticide evaluatory procedures. Because of the limitations of traditional field trials in testing all soil and climate effects on the fate of new pesticides, interest has developed in complementing these experiments with computer simulation of chemical movement through the soil profile. Such an approach can be useful in identifying which pesticides are most likely to leach to ground water and therefore require the closest scrutiny before widespread use. The Budget Approach All of any pesticide that is added to an ecosystem is removed by transport, chemically converted into other substances, or built up in storage. For a given ecosystem, such as the top foot of soil in a potato field, we can describe a "budget" which itemizes the sizes of the source, fates, and net buildup of a given chemical over some period of time. Figure VII-1 illustrates a generic model of pesticide behavior, and Table VII-1 shows the budget in corresponding tabular form. An evaluated budget for a given combination of input, management practices, anal environmental factors can be tested for "safety" by scrutinizing specific terms that are associated with environmental standards. For example, if we specify that the concentration of a substance may be no greater than 10 ug/1 in recharge, as an annual average, we can calculate an annual "allowable load" in recharge by multiplying this concentration times the volume of recharge water. If we have 50 acre- inches of recharge water, the Application, fallout ~ Dissolution Decomposition, volatilization Exudation Runoff and leaching FIGURE VII-1. GENERIC MODEL OF PESTICIDE BEHAVIOR IN THE ROOT ZONE, EXPLANATION Flow Storage corresponding allowable load would be 10 ug/1 times 50 acre-inches, totalling 0.11 lb of the substance per year. This figure can be compared to the budget's "leaching" term to assess whether there could be a problem. Budgets are constructed using a combination of field data and mathematical models that synthesize the data into a coherent whole. The following sections discuss how such models can be developed interactively with field experimentation. Bttil~i.~o Mathematical Models into the Pesticide Registration Process Models improve the development and registration process in five ways by: · Forcing completeness of investigations; Yielding insight into the relative importance of different processes affect- ing the fate of the applied chemical; · Providing explicit estimates of uncertainty; Determining the impor- tance of uncertainties relative to pending decisions; and Producing standardized, objective, and auditable conclusions. To achieve these goals, model development and use need to be incorporated into the earliest stages of the pesticide review process, as outlined in Figure VII-Z. 57 Guidin~ Field lAvest~ations Field investigations usually reflect the perceptions and biases of the investigator about which processes are most significant and which are least well known. The most significant and uncertain processes according to this conceptual model then receive the most intensive observation. Early use of an objective mathematical model can help to force completeness of the field work. Since many processes affect the fate of a pesticide in the field, it is easy to omit elements of the soil-water system which subsequently may be found to have an important role. TabLe Undissolved lb/A]¥r [lq: Application --- OUT: Dissolution --- In Soil Solution Ib/A/yr Mineralization --- Root exudation D esorptlon --- OUT: Decomposition --- Volatilization --- Leaching --- Runoff --- Uptake by pla~ts --- Immobilization --- Adsorption --- CHANGE IN STORAGE: --- Adsorbed onto Soil lb/A/yr E~: Adsorption --- OUT: Desorption Decomposition --- CHANGE IN STORAGE: --- 58 alternative mathematical models FIGURE VII-2 INTEGRATION OF PATHEMATICAL MODELS INTO THE PESTICIDE EVALUATION PROCESS, An objective mathematical model of the field processes also can help to determine which of the processes deserve closer scrutiny. Initially~ laboratory experiments~ technical literature, and other indirect sources can be used to create a formative data base. Then, the model can be used to conduct a sensitivity analysis within the plausible range of the parameters. The range of the resulting budgets will yield estimates of uncertainty and significance for each term. Figure VH-3 shows the result of a simple sensitivity analysis for the fates of aldicarb. In such a simplified system, only a few parameters need to be varied to obtain a feeling for the overall uncertainty. Here, climate~ plant uptake, and microbial decomposition parameters were varied. Despite inadequate field data for these parameters~ the sensitivity analyses show that even under very favorable assumptions~ about a third of the pesticide can be expected to leach. This amount would lead to a concentration in ground water considerably higher than the 7 ppb N.Y.S. gnideline for drinking water, so an analysis of this type could have been used to predict and prevent Long Island's current aldlcarb problems. Table VI[-Z inventories the data requirements of an early-stage model. Considerable understanding can be gained using only existing data sources. 1.0 Figure VII-3 Sensitivity of Annual Aldicarb Mass Balances to Decomposition and Plant Uptake Parameters 0.0 0.0 1.0 Leached decay rate 0.006 (1/day) plant uptake par~eter = 0.0 Leached Leached Leached Taken up by Plante Leached Taken up by Plants Taken up by plants 0.0 decay rate 0.006 0.0 d~cay rate 0.006 0.0 decay rate 0.006 (1/day) (1/day} (1/dav) plant uptake parameter = 0.067 Leached plant uptake parameter = 0.033 Leached plant uptake parameter Leached Taken up by Plante Taken up Taken 0.0 i i O.0 0.006 0.0 0.006 0.0 0.006 O.0 0.006 6O T~ble VII-2 Crop ~nd soil management SATURATED ZONE Water Movement Geologic data Permeability Improving Budget Estimates During Field The range of uncertainty included in the early sensitivity analysis will narrow considerably, if the field work has been designed and conducted well. As field results become available, the runge can be reevaluated. At this point, the pesticide's potential environmental impact can be compared with some environmental guideline such as the drinking water guideline. If initial field experiments show a clear-cut acceptable or 61 unacceptable impact regardless of the degree of uncertainty, then an early decision can be made. If the range of estimated impact still includes both possibilities, further field experi- mentation would be necessary to reduce the uncertainty until a clear-cut case one way or the other is achieved. Figure V~-4 illustrates these examples, showing when a particular pesticide can be accepted or rejected for widespread use and when further analysis is needed. Such a "multi-track" scheme would save time and avoid unnecessary expenses by focusing detailed analyses ~ ~ ~mpact REJECT WITHOUT FURTHER FIELD WORK B Time C ne D Time Time ACCEPT WITHOUT FURTHER FIELD WORK ACCEPT AFTER SUBSEQUENT FIELD WORK REJECT AFTER SUBSEQUENT FIELD WORK range of estimated impact FIgUrE Vll-q. RELATIONSHIPS AMON6 IMPACT LIMITS, UNCERTAINTIES, AND DECISIONS ABOUT PESTICIDES FOR POTENTIAL USE IN SOUTHOLD, N,Y, primarily on those chemicals for which the environmental impacts are most unclear. The pesticide registration process could be conducted more rapidly, and Rround-water quality could be more effectively safeRuarded because field, laboratory~ and computer analyses could be directed toward those pesticides for which the leaching potential is most uncertain. 62 Figure VII-5 outlines this proposed review procedure for pesticides, showing how a mathematical model based on existing field and laboratory data can be used to determine when a chemical can be accepted or rejected, or when further analysis is needed. Select spatial and temporal units, and formulate experimental design Specify water standards or criteria Estimate total recharge water under design climate for each unit of space and time ! , [Compute maximum allowable pesticide load consistent with criteria or standard ! . erive permissible loading allowing '~for uncert~inities and extreme events Compile data and formulate model for the proposed pesticide management practice Reformulate Are model and data and remedy data ~ate to he hypothe$i~ requirements Yes NO Accept pesticide in terms DJ roundwater risks Yes Yes Is an for pesticides possible? No Reject pesticide for proposed use FIGURE VII-5. PROPOSED REVIEW PROCEDURE FOR PESTICIDES, ASSUMING THE NULL HYPOTHESIS THAT THE PESTICIDE WILL NOT VIOLATE GROUND-WATER CRITERIA OR STANDARDS, 63 Chapter VIII CONCLUSIONS AND RECOMMENDATIONS Ground-water Contamination in Southold Ground water in Southold is an essential resource in need of protection and reclamation from contamination by nitrates, pesticides, and organic chemicals from residential, agricultural, and other sources. Nitrate and aldicarb contamination are widespread on the North Fork, and other pesticides and organic compounds have also been detected in some of the Town's wells. Restoration of the aquifer depends on protection of the quality of ground-water recharge since this recharge water will gradually ~flush out' and replace the water that is in the aquifer now. Long term protection of Southold's aquifer will require controlling a large range of contaminants including nitrate, pesticides, and the many organic compounds used by industry, commerce, and residential households. Nitrates Nitrate contamination of ground water comes from two major sources: (1) residential development~ including septic systems and lawn fertilizers, and fertilization of agricultural crops. In the case of residential development, nitrate control involves minimizing the use of lawn fertilizers and keeping housing densities low enough to spread out the impacts of lawns and septic systems. As shown in Figure IH-3~ simulated average nitrate recharge concentrations for Southold vary from less than two to 10 rog/1 as housing densities range from one- half to four houses per acre and as lawn fertilization rates range from zero to 3 lb N/1000 sq. ft. Consideration is being given in Southold to clustering of residential developments in order to preserve open space and valuable agricultural land. If housing is allowed to be clustered, then the resulting open space should be reserved for unfertilized vegetation or other uses which would provide the highest quality recharge water. Housing and agriculture are both intensive uses contributing nitrate and other contaminants to Southold's ground water. Lower density housing would be a more compatible use with most forms of agriculture than would high-density clustered developments. When housing is clustered, the remainder of the acreage within the development site should be reserved for unfertilized woodland in order to minimize its contribution of contaminants to the ground water. Lands that previously have been cleared for agriculture or other purposes can be revegetated using plants suitable to the local climateological and environmental conditions. In clustered developments~ this revegetation should be the responsibility of the developer. Cooperative Extension, the Soil Conservation Service, and the Nonpoint Source Handbook in preparation by the Long Island Regional Phmning Board (LIRPB, 1983) can provide guidelines for revegetation of cleared areas using low maintenance varieties suitable for the climate and soils in Southold. Farm land in Southold is a valuable resource which forms a basis for the Town's economy and rural character. In order to make farming more compatible with ground-water quality, efforts have focused in recent years on devising fertilization schemes which will maintain optimum crop yields while reducing leaching losses to ground water. Research in the 1970's on Long Island potato farming showed that fertilization rates at that time were generally higher than needed, resulting in excessive losses to ground water. Farmers subsequently have cut down on their fertilizer nser so leaching losses also have been reduced. Field studies and the computer simulations presented in this report show 150 lb N/A to be a good average rate for potatoes, with needs differing slightly between varieties. Splitting the application so that some of the fertilizer is applied six to eight weeks after planting generally results in lower leaching losses and higher uptake by the potato plants than if all is applied at the time of planting. Other crops have not been researched as thoroughly but will need to be as they become more widespread in Southold. Pesticides Pesticides in the past have not been adequately regulated for protection of ground water on Long Island. Now that aldicarb, carbofuran, and several other pesticides have been detected in many Long Island wells, much more careful attention is being devoted to assessment of the leaching potential of proposed chemicals. The pesticide review and registration process has become so cumbersome and expensive~ however, that chemical companies are reluctant to develop new products unless they will have a very large market. Development of chemicals specific to the Colorado potato beetle and useful in Long Island's Integrated Pest Management (~PM) programs therefore would probably not be given a high priority because of the high expense for a relatively limited market. What is needed is an effective technique for screening potential pesticides so that those with very low potential for leaching to ground water can proceed more quickly through the registration process, and those for which the environmental effects are less clear would receive more detailed study and review. B. esearch and education are being carried out on ways of reducing the dependency of both farmers and homeowners on chemical pesticides. This is an important goal everywhere, but particularly on Long Island, where these chemicals can easily become serious ground-water contaminants. Increased emphasis needs to be given to developing means such as biological controls and cultural practices which can reduce pest pressures without impairing environ- mental quality. Organic Chemicals Synthetic organic compounds have a large and ever-growing list of uses in industrial, commercial, and household products. Some of these compounds are highly toxic, and the long-term effects of ingesting low levels are unknown. Discovery of these chemicals in ground water is of particular concern because their sources are so diverse and widespread that detection and control are difficult problems. The Suffolk County baa on organic septic tank and cesspool cleaners is a step in the right direction for controlling these products. More study is needed on the amounts and types of compounds used in residential, commercial~ and industrial operations and on their potential for leaching to ground water. Disposal of hazardous chemicals, including empty containers~ also is a diverse and difficult problems but one of pressing urgency for ground-water protection. A study of existing data on ground- water contamination by organic chemicals on Long Island showed the percentage of wells affected to be proportional to housing density in four residential communities (Figure rrr-9). Maintaining low housing densities is one step the Town of Southold could take to limit the potential for organic chemical contamination from household products and from the associated commercial services that would accompany increases in housing and population. 65 Recommendations for Action Ground-water protection in Southold will require responsible land use planning as well as reduced dependency of farmers and homeowners on fertilizers, pesticides, and various other organic chemical products. Large scale educational efforts are needed to inform the public of the link between chemical use and contamination of the underlying aquifer and to present alternative methods for fertilization and for pest and disease control. The following specific actions would help to minimize future contamination of Southold's ground water. A. Land Use 1. Agricultural Land Preservation. One of the major issues facing the Town is how to preserve the valuable agricultural lands that form a basis for the area's economy and scenic rural character. One important step is to establish a program for preservation of prime agricultural lands. Another is to continue striving for agricultural practices that produce high quality ground-water recharge, as discussed in the Agricultural Management and Research sections below. One or fewer houses per acre would meet this guideline at lawn fertilization rates up to 3 lb N/1000 sq. ft. At lower lawn fertilization rates, other densities would also be acceptable, as shown in Table rff-4. If the Town decided to plan for compliance with the 10 rog/1 standard 99 rather than 90 percent of the time, then 1-acre zoning would be acceptable only if homeowners refrained from using any lawn fertilizers. Otherwise, lower density housing would be necessary (Table ][II-4). * Lot sizes in clustered developments should be required to conform to these same criteria. Land use planning based on nitrogen loadings will not necessarily provide adequate protection from contamination of the ground water by organic chemicals. Use of these chemicals is so diverse and widespread that sufficient data are not available for devising loading rates associated with different housing densities similar to those for nitrogen. In general, however, limiting residential development to the lowest feasible density would minimize the number of potential sources of organic contaminants from houses and their support services. Z. Housing Development. Another major land use issue in Southold is housing development: where it should be located and in what densities. One potential ground-water problem caused by housing is nitrate contamination of recharge from septic systems and lawn fertilizers. In order to plato residential development compatible with ground-water quality, the Town should choose a planning standard for nitrate based on the New York State drinking water standard of 10 mg/l. For example, if the Town decides that individual water samples should comply with the 10 rog/1 standard 90 percent of the time, then a 6 rog/1 guideline would be chosen for planning purposes. 3. Ground-Water Management Plan. Southold Town should develop a ground-water management plan that would deal with two basic problems: (1) the control of pollution sources in order to prevent the discharge of contaminants into groundwater; and (Z) the management of ground-water which is already contaminated. The plan should locate critical recharge areas, identify major point and nonpoint sources of pollution, and locate areas with sufficient high quality water to meet existing and projected future needs. Southold should adopt land use ordinances aimed at protection of These conclusions are based on the lawn sizes listed in Table II-Z and on the assumption that 50% of the grass clippings are removed from the lawns and used in ways not contributing to nitrate leaching. designated watershed areas. These ordinances might include any or all of the following provisions: -provide for maintaining open space - restrict clearing of natural vegetation restrict amount of turf (and other ornamental vegetation requiring fertilization) associated with residential, commercial and industrial development - regulate fertilizer and pesticide use - restrict density of residential development - prohibit stockpiling of any farm animal waste regulate the storage, han~lling, and use of hazardous materials (including pesticides, industrial solvents, certain household products, etc.) prohibit operation or location of any refuse disposal area. B. Agricultural Management 1. Fertilization. The fertilization of potatoes should be limited to 150 to 175 lb N/A/yr depending on variety. This total rate of application should be split so that a portion is applied at planting and the remainder is applied approximately at the two to eight inch stage of growth, when the plants are growing rapidly and have a better chance of taking up the nitrogen before it is leached. Vegetable fertilization practices also should be designed to maximize plant use efficiency and minimize loss of nitrate to ground water. l. Alternative Cropping Systems. Southold farmers should give consideration to crop diversification and crop rotations. These practices will reduce the need for pesticides by reducing insect populations and by slowing the insects' development of resistance to pesticides (Wright et al.~ 1983). With the proper choice of alternate crops, farm income levels can be maintained (Lazarus and White, 1981). 3. Education. Cooperative Extension should continue and expand their efforts to inform the agricultural community of new and improved management and cultural practices. Specifically, educational materials and programs should stress the following: a. Recommend fertiliza- tion of potatoes at 150 to 175 lb N/A/yr, depending on variety, with specific rates given for each variety whenever possible. Promote vegetable fertilization practices which will minimize leaching of nitrate to ground b. Encourage potato farmers to split the fertilizer application, as specified above (see Fertilization). c. Emphasize the benefits of crop diversification and crop rotation as means of reducing pest pressure, lowering pesticide use (and assocaited costs), and rejuvenating soils. d. Demonstrate cultural and biological methods of pest and disease control, and encourage their widespread use wherever feasible. As new guidelines are developed for all crops (see Research below), educational materials should be made available and plots established to demonstrate the new techniques. C. Residential mud Commercial Land Management 1. Lawn Fertilization. Low maintenance turf varieties should be used for residential and commercial lawns, thus reducing the need for fertilization. For existing lawns which are not of a low maintenance variety, fertilization rates should not exceed 2 lb/N/1000 sq. ft./yr. Grass clippings should generally not be removed, but should remain on the lawns to serve as a slow-release source of nitrogen fertilizer. If grass clippings are removed, they should be utilized as mulch or compost and reapplied to lawn or garden areas in order to reduce the need for commercial fertilizers. Z. Chemical Use and Disposal. The use of chemical pesticides and herbicides should be minimized, with emphasis given to alternative (non- chemical) means of combating common pests and diseases. Disposal of potentially hazardous materials such as paint thinners, used motor oil, cleaning solvents, etc., should be accomplished in such a way that they do not present a threat to ground water. Disposal of such materials into cesspool systems or directly onto the ground are not acceptable methods, since they will result in ground-water contamination. 3. Education. The public awareness of the link between chemical use and ground-water contamination must be heightened. Various public agencies, including Suffolk County Cooperative Extension (S.C.C.E.), Suffolk County Department of Health Services (S.C.D.H.S.), New York State Department of Environmental Conservation (N.Y.S. D.E.C.), and the Town of Southold should prepare and distribute educational materials which indicate the threat to ground water (and ultimately to drinking water) presented by fertilizers, pesticides and other organic chemicals, and which encourage minimal use of all such 67 substances. Specifically, Cooperative Extension should recommend keeping lawn fertilization rates below 2 lb N/1000 sq. ft./yr., leaving grass clippings on the lawns, and using low maintenance turf varieties on all new lawn areas. Public education also should address proper means of disposing of chemicals commonly used in residences and commercial establishments. It is imperative that the public be made aware of the serious threat to ground-water quality posed by many common household chemicals. A substantial educational effort will be necessary to reduce public reliance on chemicals and should be stressed in all pertinent publications, news releases, workshops, etc., produced by S.C.C.E., S.C.D.H.S., N.Y.S.D.E.C. and the Town of Southold. D. Research 1. Agricultural Practices. The Long Island Horticultural Research Laboratory (L.I.H.R.L.) should expand and emphasize its research into and demonstrations of techniques for reducing agricultural pesticide and fertilizer use. a. The Integrated Pest Management Program and associated development of biological and cultural practices for pest control should receive top priority in work by the L.I.H.R.L. b. Nitrogen fertilization rates specific for Long Island conditions should be developed for all crops to maximize nitrogen use efficiency in order to minimize nitrate leaching to ground water. c. Monitoring of ground water for currently used pesticides and experimental compounds should be continued and expanded. d. Techniques should be developed to increase the feasibility of crop rotations and alternative crops. Research should include economic evaluations of labor and equipment needs as well as marketing potential. Z. Organic Contaminants. In an area with septic systems and highly permeable soils, leaching of household products containing toxic organic compounds is a concern. Surveys on the types and amounts of these products in industrial, commercial, and household use would provide a start in determining their loading rates to ground water. Monitoring for these chemicals by the Suffolk County Department of Health Services should be continued and expanded to provide an early warning of contamination problems. 3. Aldicarb Simulation. Another area where more research is needed is in predicting the timing of aldicarb contamination at individual wells in Southold- Cornell's computer model in its present form can trace the timing and aerial extent of aldicarb movement through the aquifer but is not designed to accurately predict when individual wells will be affected. This refinement of the model would be useful in designing monitoring and water treatment programs because it would identify the areas where contamination would be expected to be most severe at any particular point in time. 4. Pesticide Screening. A mathematical model based on existing field and laboratory data should be developed for utilization by regulatory agencies (USEPA and NYSDEC) in screening pesticides. Such a screening would indicate where additional field and laboratory research are required~ prior to the registration of pesticides for use on Long Island. E. Regulation and Enforcement 1. Turf Control. The Town of Southold should consider the adoption of a turf control ordinanc% to limit the amount of tu~f associated with future residential and commercial development. Limitations could take the form of maximum turf area per home or maximum percentage of each lot~ and encouragement of low maintenance practices. Z. Pesticide Screening. The U.S. Environmental Protection Agency~ N.Y.S. Department of Environmental Conservation~ and Suffolk County Department of Health Services should carefully screen any proposed agricultural chemicals to identify their potential for leaching to ground water under conditions typical of Long Island. Continued research and field monitoring could then focus primarily on those chemicals identified to be potential ground-water problems~ and other chemicals with a lower risk could be moved more quickly through the registration process. 3. Consumer Product Bans. The Suffolk County ban on septic system cleaners should be strictly enforced because these products add organic pollutants to the ground-water. If other industrial or household products are identified as ground-water contaminants, prohibition of their use should also be considered at the County level. 4. Interagency Coordination. The effectiveness of existing surveillance and enforcement personnel should be enhanced through improved information exchange. The NYSDEC and SCDHS should jointly prepare information handbooks or pamphlets explaining the proper storage~ handling~ and disposal of hazardous materials. This informational material should be distributed to the town building department~ town fire inspectors~ local fire departments, and other agencies whose routine inspections or approval activities might uncover or prevent improper use or disposal of hazardous materials. household Similar information on the use of hazardous materials 69 should be prepared and disseminated to the general public. 70 ?l Chafer IX Pi)SI'SCRIPT In many parts of the country~ ground water is taken for granted and assumed to be invulnerable from contamination by human activities. On Long Island~ in contrast, discovery of pesticides~ synthetic organic compounds~ nitrate~ and other pollutants in the aquifers supplying the area~s sole source of drinking water has raised ground-water quality to an issue of urgent concern. As a result~ much attention has focused on means of providing potable water to meet existing and future needs. A more difficult~ but equally urgent problem is how to protect the quality of water recharging the aquifers since it is this recharge water which will replenish and restore the contaminated supplies. The work presented in this report was designed to provide information useful in making laud use and management decisions consistent with protection of recharge quality in Southold. Recommendations were made concerning land use pl~ning~ residential development~ agricultural managements public education, research~ and regulations that would help to restore and protect the area's ground-water quality. For Southold this is an important time for water quality planning~ the Town currently is revising its land use master plan~ and the County recently has completed an assessment of water supply options to meet existing and future needs. The opportunity is ripe for developing a Town ground-water management plan to incorporate recharge quality as an integral consideration when land use or management decisions are made. The ground-water evaluations and recommendations presented in this report are intended both to assist Southold in managing its ground-water resources and to provide an example to other communities of how ground-water considerations can be integrated into land use planning and management processes. REFERENCES Baler, J.H., and S.F. Robbins. 198Z. Report on the Occurrence and Movement of Agricultural Chemicals in Groundwater: North Fork of Suffolk County. Suffolk County Department of Health Services, Bureau of Water Baier, J.H., and K.A. Rykbost. 1976. The Contribution of Fertilizer to the Groundwater on Long Island. Paper presented at the Third National Groundwater Quality Symposium, September 15-17, 1976. Bond, M.C. 1947. Suffolk County A~riculture and Land Use. Department of Agricultural Economics, A.E. 634. New York State College of Agriculture, Cornel1 University, Ithaca, New York. Chapman, R.A., and C.M. Cole. 198Z. Observations on the Influence of Water and Soil pH on the Persistence of Insecticides. $. Environ. Sci. Health B17(5):487-504. Chu, C.C., amd G.W. Selleck. 1977. A Survey of Commercial Potato Fertilization Practices on L.L~ 1975-1977. Suffolk County Agricultural News, Vol. LX1, October 1977. Cooperative Extension Association of Suffolk County. 1981.* Home Horticultural Facts: Lawn Hansen, J.L. and M.H. Spiegel. 1983. Long Term Hydrolysis Studies of Aldicarb Sulfoxide and Aldicarb Sulfone. Union Carbide Corp., Unpublished Report. Hansen, J.L., R.R Romine and M.H. Spiegel. 198Z. Hydrolysis studies of aldicarb~ aldicarb sulfoxide and aldicarb sulfone. Environmental Toxicology and Chemistry Z:147- 153. Intera Environmental Consultants, Inc. 1980. Mathematical Simulation of Aldicarb Behavior on Long Island: Unsaturated Flow and Ground-Water Transport. Houston, Texas. Kain, D.P. and T.S. Steenhuis. 1983. Adsorption Partition Coefficients and Degradation Rate Constants and Half-Lifes of Selected Pesticides Compiled from the Literature. Unpublished Manuscript. Cornell University, Ithaca, New York. Kossack, R.S., and G.W. Selleck. 1979. Potato Yields~ Fertilizer Use and Water Pollution on L.L Suffolk County Agricultural News, Vol. LXI]I, March 1979. Lazarus, S.S., and G.B. White. 1983. The Economic Potential of Crop Rotations in Long Island Potato Production. Department of Agricultural Economics, Cornell University, Ithaca, New York. Lemley, A.T. 1983. Unpublished Results. Department of Design and Environmental Analysis, Cornell University, Ithaca, N.Y. Fertilization. Riverhead, New York. 73 Long Island Regional Planning Board. 1983. Z08 Areawide hnplementation Plan - Nonpoint Source Handbook. In press~ Hauppauge, N.Y. Meisinger, J.J. 1976. Nitrogen Application Rate Consistent With Environmental Constraints for Potatoes on Long Island. Search Agriculture, Vol. 6, No. 7. Cornell University Agricultural Experiment Station, Ithaca, New York. National Research Council. 1977. Drinking Water and Health. National Academy of Sciences. Washington, D.C. Pacenka, S. and K.S. Porter. 1981. Preliminary Regional Assessment of the Environmental Fate of the Potato Pesticid% Aldicarb~ Eastern Long Island~ New York. Center for Environmental Research, Cornell University, Ithaca, New York. Pike, $., S. Goldfarb, and K.S. Porter. 1980. 1980 Survey of Turf Management Practices in Nassau and Suffolk Counties~ L.I. Center for Environmental Research, Cornel1 University, Ithaca, New York. Porter, K.S. 1982. Groundwater Information: Allocation and Data Needs. Center for Environmental Research, Cornell University, Ithaca, New York. Porter, K.S. and N. Beyer. 1977. Report on Aldicarb. Suffolk County Cooperative Extension Association, Riverhead, New York. Rykbost, K.A., P.A. Schippers, C. Chu, G.W. Selleck, D.R. Bouldin, 5.L. Ozbun, G. Rathbun, and R.S. Kossack. 1979. Fertilization of Potatoes as a Source of Ground Water Contamination. Appendix I, 197_ Annual Report of the Long Island Horticultural Research Laboratory, Riverhead, New York. Rykbost, K.A., P.A. Schippers, R.S. Kossack, G.W. Selleck, C.C. Chu, and D. Bouldin. 1979. Studies in Fertility and Nitrate Pollution in Potatoes and Turf on Long Island. Long Island Horticultural Research Laboratory, Riverhead, New York. Selleck, G.W., R.S. Kossack, C.C. Chu, and K.A. Rykbost. 1980. Studies on Fertility and Nitrate Pollution in Turf on Long Island. Long Island Horticultural Research Laboratory, Riverhead, N.Y. Snow, J.T. 1976. The Influence of Nitrate Rate and Application Frequency and Clipping Removal on Nitrogen Accumulation in a Kentucky Bluegrass Turf. M.S. Thesis, Cornel1 University, Ithaca, New York. Snyder, D.P. 1981. Cost of Production: Update for 1981. A.E. Res. 8Z-Z0. Department of Agricultural Economics, Cornell University, Ithaca, New York. Snyder, D.P. 1977. Cost of Production: Update for 1976. A.E. Res. 77-11. Department of Agricultural Economics, Cornell University, Ithaca, New York. Tingey, W. 198Z. Report presented to the College of Life Sciences and Agriculture. November 198Z. Cornell University, Ithaca, New York. Underwood, F.L. 1933. Costs and Returns in Producing Potatoes in New York in 19Z9. Bulletin of the Cornell University Agricultural Experiment Station. U.S. Department of Agriculture. 1954. Potato Production in the Northeastern and North Central States. Farmers' Bulletin No. 1958. 74 Wilkinson, C.F., J.G. Babish, A.T. Lemley, amd D.M. Soderland. 1983. A Toxicological Evaluation of Aldica~-~ and its Metabolites in Relation to the Potential Human Health Impact of Aldicarb Residues in Long Island Ground Water. Institute for Comparative and Environmental Toxicology, Cornell University, Ithaca, New York. Wright, R.J., R. Loria, J.B. Sieczka, and D.D. Moyer. 1983. Final Report of the 198Z Long Island Potato Integrated Pest Management Plot Program. V.C. Mimeo Z87. Department of Vegetable Crops. L.I. Horticultural Research Laboratory, Riverhead, New York. Appendix A Potato Pest Management on Lon~ Island Potatoes have been a major crop on Long Island for the past 100 years, and the monoculture nature of the potato farming has led to problems both for pest control and for ground-water quality. Pest populations have built up, requiring large doses of pesticides, which in turn have led to loss of effectiveness of various chemicals because of development of resistant populations. When aldicarb (brandname Temik) was used from 1975 to 1979 on Long Island, it proved to be highly effective in controlling the Colorado potato beetle and nematode populations. It had to be banned, however, because it readily leached through the sandy, acidic soils and was found in ground water in concentrations far exceeding the 7 ppb New York State drinking water guideline. Since aldicarb became unavailable, Long Island growers have increased both the number of pesticide applications and the total amounts applied, resulting in higher costs but not necessarily comparable crop protection. There is now a critical need to develop effective techniques to control pests if potato farmers are to sustain adequate production and remain in business. At the same time, further serious contamination of ground water used for drinking water by new application of pesticides is unacceptable to all residents of the North Fork. Given this double constraint, the need to control pests, and the need to protect ground water quality, Cornell University and its Long Island Horticultural Research Laboratory have instituted a vigorous program of research to discover new methods of pest controls for Long Island potato farmers. This section briefly highlights some of this work. For more detail, the reader is referred to reports prepared by staff of the Long Island Horticultural Research Laboratory. Two overall types of methods for managing potato pests can be distinguished: Direct pest control techniques for use in potato production. B. Alternate cropping systems. The work now being pursued is evaluating both conventional and innovative approaches to combating the pests in both types of methods. A. Direct Pest Control Techniques Marketable yields of potatoes on Long Island can be severely limited by the following pests: · Colorado potato beetle · Golden and meadow nematodes · Potato scab · Weeds Other pests, such asaphids, canalso cause serious damage if uncontrolled, but there are effective control practices available to farmers. 1. Colorado Potato Beetle The Colorado potato beetle on Long Island has developed into a voracious pest capable of rapidly becoming resistant to pesticides used against it. Since the development of resistance may be in part directly related to the degree of exposure the beetle has to a pesticide, one way of delaying resistance would be to reduce the frequency and amount of use of particular pesticides. Staff at the Long Island Horticultural Research Laboratory are formulating acti(m thresholds for the beetle below which damage caused by the insect may be tolerated and spraying is thereby unnecessary. Initial work with farmers has been encouraging and suggests that a reduction in the use of pesticides is a practical option in some circumstances (Wright et al., 1983). Cultural practices which do not depend on pesticides for reducing the population of the Colorado potato beetle are also being evaluated. For example, earlier applications of chemicals intended to kill the potato vines prior to harvest could beneficially reduce the food available to the beetles. Biological controls of the beetle are also being studied although these options are unlikely to become available in the near future. One promising natural control is that achieved by a fungus called Beauveria bassiana. The fungus is completely safe in environmental and ecological respects and may be effective in at least suppressing the numbers of beetles that may develop each season. Another promising but long term biological option is to develop potato plants which are resistant to the beetle. Some wild species of potatoes, which are known to be resistant, are now being interbred with domestic varieties. One type of resistance is provided by the presence of substances called glycoalkaloids in the potato plant. Unfortunately, the substances lack discrimination in which parts of the potato plant they are found, and hence occur in the tubers as well as the leaves. Another defensive mechanism found in some potato species are adhesive or sticky hairs. When touched by an insect, these hairs exude a viscous ~um which sticks to the insect and rapidly hardens (Tingey, 198Z). The insect is eventually immobilized and dies. Affected insects sometimes assume a ~Charlie Chaplinn appearance with large gum boots on each leg (Figure A-IL The sticky hairs are effective against aphids as well as the Colorado potato beetle. FIGURE A-l, AN APHID ON THE LEAF OF A HYBRID POTATO PLANT THAT PRODUCES A STICKY BLACK GUM, TRAPPING INSECTS AND CAUSING THEM TO STARVE TO DEATH, Z. Nematodes The pesticides now available for nematode control (nematicides) are of limited and variable effectiveness. A more immediately promising option is to introduce potato varieties which are resistant to the nematode. However, the existing resistant varieties may lack the qualities or characteristics which would lead to their adoption by Long Island farmers. Short-term options to control the golden nematode are therefore seriously limited. 3. Potato Scab Potato Scab (Figure A-Z) is especially significant for farmers on Long Island because the most effective method of control is for the farmers to m~taln a low pH on their potato fields. Unfortunately, crops which might otherwise be used in a rotation with the potatoes become impractical possibilities because they would be, to some de~ree, incompatible with the high acidity in the soil. Dealing with potato scab on Long A-3 FIGURE A-2, A POTATO SHOWING THE EFFECTS OF THE POTATO SCA~ DISEASE. Island has, therefore, fostered the monocultural practices which in turn make it more difficult to deal with potato pests such as the Colorado potato beetle and the golden nematode. Currently, work at the Long Island Horticultural Research Laboratory is investigating the biolo~w and ecology of scab. Scab resistant potato varieties are available, but these are not considered to be as commercially attractive as potatoes now commonly grown on Long Island. 4. Weeds Weeds which invade potato fields compete with the crop for space, nutrients and water. The weeds may also harbor pests as well as hindering the efficiency with which the potato tubers can be harvested. Weed control is therefore a serious economic issue for the potato farmer. Two primary weeds are barnyard grass and nutsedge (Figures A-3 and A-4). Bindweed (Figure A-5) has shown an increasing tendency to mitrate from its traditional home, the roadside, and is impacting potato production in some areas. There currently are no practical controls for these weed species, and an overall control strategy needs to be developed incorporating information on their biology and ecology as well as whatever chemical and cultural practices are available. FIGURE A-3. BARNYARD GRASS (ECHINOCMLOA CRUS- GALLI) A COMMON WEED IN POTATO FIELDS IN SOUTHOLD~ N.Y, FIGURE A-q, NIJTSEDGE (CVpERUS ROTUDUS), A COMMON WEED IN POTATO FIELDS IN SOUTNOLD, N,Y, FIGURE A-5. B!NDWEED (CONVOLVULUS ARVENSIS), A WEED COMMONLY POUND ALONG ROAD- SIDES WHICH ALSO INVADES POTATO FIELDS IN SOUTHOLD, N.Y. B. Crop Rotations and Alternative Crops Crop rotation can break the life cycles of some pests, thereby decreasing the need for pesticides and slowing the development of insect resistance to each chemical. Rotation also helps to maintain soil vitality because each crop depletes or restores soil nutrients in differing ways. For many years, potato farmers have followed a seasonal rotation by sowing rye after the potatoes have been harvested. The rye is plowed under prior to planting the potato seeds in the following spring and adds to the soil organic matter. To the extent the rye increases soil biological activity in the soil, it may assist in the retention and biodegradation of pesticides. The rye cover also decreases wind erosion during the winter months. 1. Annual Crop Rotations However, for reasons already discussed, annual rotations are not followed to an extent which could break the pattern of monocultural production. Recent work carried out at Cornell and at the Long Island Horticultural Research Laboratory suggests that crop rotation might nevertheless be beneficial. For example, Wright et al. {1083) report that on three farms out of four, potatoes grown on fields previously used for another crop had lower populations o2 the Colorado potato beetle than did fields on which no rotation took place. On the fourth farm, the overall number of the Colorado potato beetle were so low that differences between the rotated and non-rotated fields were immaterial. A preliminary economic evaluation carried out by Lazarus and White (1985) showed that rotation of some potato fields with cauliflower would restflt in decreased pesticide use and greater income than if all fields were in continuous potato production. Most other rotations decreased net returns to some extent, so crops would need to be carefully chosen to maintain farm income levels. The adaptability of potato farmers is, however, restrained by the high degree of specialization they have achieved. The investment in equipment and facilities such as storage sheds represent a major commitment of capital that has limited transferability to other crops. Z. Housing As An Alternative An obviously profitable alternative "crop" for many former potato farmers on Long Island has been housing. From an environmental viewpoint, the replacement of farms with houses, highways and other land uses may be substantially more threatening than the farms they replace. Unfortunately, other crops have been less successful than residential developments in supplanting the potato. Clearly this is a consequence of the economic forces involved. However, no economic analysis has been undertaken recently which systematically evaluates the costs and benefits of these major options now A-5 confronting farmers. Over the past few years there have been some encoUraging economic tendencies: The farmland preservation program has provided economic incentives to continue farming, increasing numbers of roadside farm stands have given farmers a direct and very profitable means of marketing a wide variety of produce, and new crops such as grapes for wine production are proving to be economically viable. It therefore appears timely to undertake a macro-economic analysis of the current patterns of farm production and options. Through such an analysis it might be possible to further develop and support viable cropping alternatives in the future. B-1 RZ. Residential (Z-4 dwellings/acre} b. Assumptions applying a. to all residential Al. Agriculture: Potatoes a. (cropland in potato b. productior.) Population density 7.4 persons/acre Percentage of land Value Information Source Population density estimates were based on the 1980 Census~ which reported Allland cover percentage calculations were based on measurements from Incalinw- altitude air photos. Porter ~nd others~ 1978 NYS Department of Health Agriculture: Mixed a. Vegetables (Land b. devoted to truck crops such as tomatoes, A winter grain cover crop is assumed.} Depth of root zone Z4 inches Maximum plant nitrogen 134 lb N/A content if all growth conditions are optimal Percentage of harvested plant biomass nitrogen 66% returned to soil Percentage of harvested plant biomass nitrogen 34% removed from field Amount of inorganic nitrogen fertilizer 118 lb N/A/ye applied Irrigation water amount {Suffolk County Cooperative Extension and Jim Pike, South Fork mixed vegetable farmer. B-2 Table 15-1 (CONTINUED) Land Use Category' Assumption Valne A3, Agriculture: Vineyards (Land in table or wine grape production) a. Depth of root zone 47 inches b. Maximum plant nitrogen if all growth conditions are optimal Z7 lb N/A c. Percentage of harvested plant biomass nitrogen returned to soll 80% d. Percentage of harvested pla~t biomass nitrogen removed from field 10% e. Inorganic nitrogen fertilizer applied 30 lb N/A applied 289 lb bi/A/yr A6. Agriculture: 1st Year a. Sod (Land used for b. commercial sod farming~ with crop planted in September) c. A7, Agriculture: Znd Year a. Sod (La~d used for b. commercial sod farraing~ with crop harvested in October of the Znd year) c. Suffolk County Cooperative Extension Suffolk Cmmty Cooperative Suffolk County Cooperative Extension Suffolk County Cooperative Extension Dr. Arthur Eing~ Long Island Horticultural Research Laboratory Suffolk County Cooperative Extension Dr. Arthur Eing~ Long Island Horticultural Research Laboratory B-3 Table B-I (CONTINUED) Summary of Assumptions Relating to Each Land Use Type (1980) plant biomass nitrogen return to soil 100% Estimated from data from eastern Long Island experiments (Meisinger~ 1976) Suffolk County Cooperative Suffolk County Cooperative Extension 01. Other: Golf Courses Fairways Roughs 85% Dr. A. Martin Petrovic, 60% Cornell University Dept. Z0% of Floriculture and 3% Ornamental Horticulture Z% 10% Z73 lb N/A 100% 98 lb N/A 55 ~ N/A Z00 lb N/A ZOO lb N/A 8-4 Cover 1. Turf Assumptions (a) half of the initial content of animal waste is lost in the atmosphere. {h) all pet waste assumed to be deposited on turf. Pet popt~ation based on human population ~ud dogs a~d cats per person for Long Lsland. Pet waste production is about 5,3 grams lq per daF per dog and 4 grams N per day per cat Turfgrass leaf density is assumed to vary during the ye~ peaking in late spring ~d summer and being least in winter. (d) Depth of root zone is gO cm (e) Turf mowings are assumed to bi-weekly during the summer and to each remove l0 percent of the biomass {f) 50 percent of harvested plant biomass nitrogen is returned to soil and 50 percent is removed. Porter (1975) and Lauer and others (1976) (1974) and Porter,.nd others (1978) Porter and others (1978) Impervious (a) There is assumed to be no root zone (b) All water is assumed to run off onto adjacent pervious Time Step Ste~ (kg/ha) Table B-3 Rates and Timings o£ Inorganic Fertilizer (Kg N/ha) Southold Demonstration Site 1st Znd Year Year Horse Potatoes Grain ColeCrops Orchards Vineyards Sod S~ Nurseries Pastures (k~ha) (kg/ha) (kg/ha) (kg/ha) (kg/ha) (k~ha) (kg/ha) (kg/ha) (kg/ha) 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0,0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0,0 0.0 0.0 0.0 0.0 0.0 0.0 0,0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0,0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0,0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 110.0 33.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 64.9 0.0 0.0 0.0 0.0 44.0 0.0 0.0 0.0 0.0 61.7 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 58.1 73.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 64.9 64.9 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0,0 0.0 0.0 0.0 0.0 0~0 0.0 0.0 0,0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 61.7 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0,0 0.0 0,0 0.0 0,0 19Z.5 0,0 154.0 110,0 33.0 1Z3.0 Z30.8 185.1 0.0 Golf Courses Mixed Rough & Veget~lesFairways Gro~ Greens Tees (kg/ha) (~/ha) (kg/ha) (~ha) (~/h~ 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0,0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 99.0 0.0 0.0 48.8 48.8 0.0 0.0 0.0 0.0 0,0 0.0 0.0 0.0 36.6 36.6 33.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 48.8 61.0 48.8 48.8 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Beginning Day of Time Turf Potatoes Time Step Step (kg/ha) (kg/ha) Jan. 1 I 0.0 0,0 Jan. 85 3 0.0 0.0 Feb. 6 4 0,0 0.0 Feb. 18 S 0.0 0.0 Mar. Z 6 0.0 0.0 Mar. 14 7 0.0 0.0 Mar. 26 8 0.0 0.0 Apr. 7 9 0.0 14,0' May I 11 89.3 0.0 May 13 1Z 0.0 0.0 Table g-4 Rates and Timings of Organic or Slow-Release Fertilizers (Kg N/ha) Southold Demonstration Site (kg/ha) (kg/ha) (kg/ha} (kg/ha) (kg/ha) (kg/ha) (kg/ha) Horse Golf Courses (Manure) Mixed Rough & Pastures Vegetables Fairways Grounds Greens Tees (kg/ha) (l~/ha} (kg/ha) (kg/ha} (kg/ha) (kg/h~ 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0,0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0,0 0.0 0,0 0.0 0.0 0.0 0.0 0.0 0.0 0,0 0.0 0,0 0,0 0.0 0,0 0,0 0.0 0.0 0.0 0.0 0.0 0,0 0.0 ZZ.O 0.0 0.0 0.0 0.0 0.0 ZZ.O 0.0 0.0 0.0 0.0 0.0 ZZ*O 0.0 0.0 0,0 0,0 0.0 ZZ.O 0,0 0.0 0.0 0.0 0.0 ZZ.O 0.0 0.0 0.0 0.0 0.0 Z2.O 0.0 Z4.4 0.0 0,0 0.0 ZZ.O 0.0 0.0 0.0 0.0 0.0 ZZ.O 0.0 0.0 0.0 0.0 0.0 ZZ.O 0.0 0.0 0.0 36.6 36~6 ZZ.O 0.0 0.0 0.0 0.0 0.0 2Z.O 0.0 0.0 0.0 0.0 0.0 28.0 0.0 0.0 0.0 0.0 0.0 ZZ.O 0.0 0.0 0.0 0.0 0.0 ZZ,O 0.0 0.0 0.0 0.0 0.0 ZZ,O 0.0 0.0 0.0 0.0 0.0 2Z.0 0.0 0.0 0.0 0.0 0.0 ZZ.O 0.0 0.0 0.0 0.0 0.0 ZZ,O 0.0 0.0 0.0 0.0 0.0 ZZ.O 0.0 0,0 0.0 0,0 0.0 0.0 0.0 0.0 0.0 48.8 48,8 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 440.0 5.1 Z4.4 0.0 85.4 85.4 Time Step Step (cra) (cra} Table Irrigation Atoou~ts and Timings Southold Demonstration Site (cra) (cra) (cra) (em) (cra) 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0,0 0.0 0.0 0.0 0.0 0.0 0.0 0,0 0o0 0.0 0,0 0.0 0.0 0,0 0,0 0.0 O.0 0.0 0,0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0,0 0.0 0,0 0.0 0.0 0,0 0.0 0.0 0.0 0,0 0,0 0.0 0.0 0.0 0,0 0.0 0,0 0.0 0.0 0.0 0,0 0,0 0.0 0.0 0.0 0.0 0.0 0.0 0,0 0.0 0,0 0.0 0,0 0,0 0.0 0,0 0,0 0.0 0,0 0,0 0,0 0,0 Z.0 0,0 0,0 0.0 0,0 0.0 0.0 3,0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0o0 0.0 ~.0 0.0 0.0 2.5 Z.0 0.0 0.0 0.0 2.5 0.0 2.5 ~.0 0.0 0.0 0.0 0,0 0,0 0.0 0.0 0.0 0.0 0.0 0.0 0,0 0.0 0.0 0.0 0.0 0,0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0,0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0,0 0,0 0.0 0.0 0.0 0.0 0.0 0,0 0.0 0.0 0,0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0,0 0.0 0.0 0.0 0.0 0.0 0.0 0,0 0.0 0.0 0,0 0.0 0.0 0.0 0.0 0,0 10,0 6.0 0.0 ~.0 8,0 Golf Courses Mixed Rough & Pastures Vegetables Fairways Grounds Greens (cra) (cra) (cra) (cra) Tees 0.0 0.0 0.0 0.0 0.0 0,0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0,0 0.0 0.0 0.0 0.0 0,0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0,0 0.0 0.0 0.0 0.0 0.0 0,0 0.0 0.0 0.0 Z,2 0.0 0.0 0.0 6,5 0.0 6,5 0.0 0.0 6.5 0,0 6,5 6.5 0,0 0.0 6,5 0.0 6.5 6.5 0.0 0.0 6,5 0.0 6.5 6.5 0.0 2.5 6,5 0.0 6.5 6,5 0.0 2,5 6.5 0.0 6.S 6,5 0.0 0.0 Z.Z 0.0 0.0 0.0 Z,Z 0.0 0,0 0.0 0.0 0.0 0.0 0.0 0.0 0,0 0.0 0.0 0.0 0.0 0,0 0,0 0.0 0.0 0.0 0,0 0.0 0,0 0,0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0,0 7.$ 60.9 0.0 60.9 60.9 Table These values range from 0.0 to 1.0, representing zero to 100 percent of the nitrogen present in a mature c~op under optimal conditions. As the plants mature, the fraction sizes increase. The fractio~s for winter cover crops never reach 1.O because they are expressed as percentages of the nitrogen potential of the summer crop rather than of the graL, m itself. Represents a winter cover crop of grain, Beginning Day of Time Time Step Step Turf Jan. I 1 0,0 Jan. 13 Z 0.0 $~. 15 3 0.0 Feb. 6 4 0.0 Feb. 18 5 0.0 Mar. I 6 0.0 Mat. 14 7 0.0 Mar. 16 8 0.0 Apr. 7 9 0~0 Apr, 19 10 0.0 May I 11 0.0 May 13 12 0.I May Z5 13 0.1 June 6 14 0.1 June 18 15 0.1 June 30 16 0.1 July 12 17 0.1 July 14 18 0. I Aug. 5 19 0.1 Aug. 17 10 0.1 Au&. 19 Z1 0.1 Sept. 10 22 0.1 Sept, 1Z 23 0.0 Oct. 4 14 0.0 Oct. 16 25 0.0 Oct. 18 16 0.0 Nov. 9 17 0.0 Nov. ZI 28 0.0 Dec. 3 19 0.0 Dec. 15 30 0.0 Table 5-7 Plant Harvest Fractions and Dates (Expressed as Percentages) Southold Demonstration Site Po/aloes Grain Cole Crops Orchards Vineyards 0.0 0.0 0.0 0.0 0,0 0.0 0,0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0,0 0.0 0.0 0.0 0.0 0.0 0,0 0.0 0.0 0.0 0.0 0.0 0.0 0,0 0,0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0,0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0,0 0,0 0,0 0,0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0,0 0,0 0.0 0.0 0.0 0.0 0,0 0,0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0,0 0.0 0,0 0.0 0.0 0.0 0.0 0.0 0,0 0,0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1st Znd Golf Courses Ye~ Yew Mixed Rough & Sod Sod Nurseries Pasture Vegetables Fairways Gro~ds Greens Tees 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0,0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0,0 0.0 0,0 0.0 0.0 0.0 0,0 0,0 0.0 0,0 0,0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0,0 0.0 0.0 0.0 0.0 0,0 0,0 0.0 0.0 0.0 0.15 0.0 O.l 0.1 O.Z 0.15 0.0 0.1 0.0 0.15 0.0 0.0 0.0 0,0 0.15 0.0 0.1 0.0 0.15 0.0 0.I 0.1 O.Z 0.15 0.0 0,1 0.0 0.15 0.0 0.1 0.1 0.Z 0,15 0.0 0.1 0.0 0.15 0.0 0.1 0.l 0.Z 0.15 0.0 0.0 0.0 0.15 0.0 O.l 0.1 O.Z 0.15 0.0 0.0 0.0 0,0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0,0 0.0 0,0 0.0 0,0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Turf Nitrogen Simulations One of the major concerns on Long Island is the contamination of ground water due to nitrates and the contribution that nitrogen fertilizers used on turf adds to this contamination. The assessment of nitrogen leaching from turf described in this report is based on a detailed simulation of water and nitrogen movement in the turf root zone using the WALRAS root zone nitrogen simulation model. The model computes the amount of nitrogen that leaches out of the root zone and gets into ground water, taking into account all of the significant environmental and management factors such as soil properties, climate, fertilization timing and rate, pet waste deposited on turf and irrigation. The details of the WALRAS simulation models can be found in manuals written by the Center for Environmental Research. This appendix describes: (1) how the model was calibrated to be consistent with field experiments on turf uptake of nitrogen, (z) a comparison of simulated nitrogen leached with observed nitrogen in ground water, and (3) an assessment of the uncertainty of the simulation results. Calibration The two most significant fates of nitrogen in the turf root zone are uptake by the plants and leaching (gaseous loss and runoffalso remove nitrogen from the root zone.) Plant uptake by turf has been extensively measured in field experiments in a way which allows a direct comparison with simulated plant uptake. Therefore we were able to calibrate the plant uptake simulation using these field results which were conducted by the Long Island Horticultural Research Laboratory, L.I.H.R.L. (Selleck et al., 1980). Between 1974 and 1976 the eight fertilization treatments described in Table C-1 were tested on experimental turf plots at the Long Island Horticulutral Research Laboratory. At the end of the experimental period the nitrogen content of the plants was measured. The amount of nitrogen fertilizer taken up by plants was determined by subtracting the nitrogen content of unfertilized tttrf on control plots from the nitrogen content of fertilized turf. These experimental conditions were simulated using the WALRAS model and the field results and the simulation results were compared. Minor adjustments were made to the parameters which control the plant uptake simulation to make the model consistent with this experiment and with nitrogen uptake results measured on sod farms as part of the same L.I.H.R.L. study. Figure C-1 shows a comparison of the simulatea and observed nitrogen uptake by turf. Line A represents the amount of nitrogen which would be in the plants if their uptake of fertilizer nitrogen was 100% efficient. Line B represents the actual amount of fertilizer nitrogen retained in the plants. The simulation results compare well with the observed results. Table C-1. Rates and dates of fertilizer N aplied to turf (Selleck and others, 1980) Treatment Fertilization Rate per Application Total N (lb N/1000 scl ft) Dates of Application /3Z mos. Total N /yr 3 4 5 6 0 .Z60 1974:5/26,6/24,9/2 9/Z4,10/9,11/5 1975 and 1976:4/1,4/Z1 5/7,6/1,9/1,9/21,10/7, 11/1 5.64 1.03 1974: 5/Z6,9/Z 1976 and 1977:4/1,9/1 6.15 · 51 as in Z (above) 11.28 1.03 1974:5/26,6/24,9/2 10/9 1975and 1976:4/1,9/1 2.05 1974: 5/Z6,9/Z 12.3 1975 and 1976:4/1,9/1 12.3 1.03 as in 2 (above) 22.55 Z.05 as in 5 (above) Z4.60 1.89 2.05 3.75 4.10 4.10 7.50 8.2 Comparison with Well Data In order to test the accuracy of the simulations for predicting ground water quality underneath turf, simulation results were compared with water quality data from shallow observation wells in the Twelve Pines subdivision in Medford, N.Y. (Figure C-Z). The subdivision was constructed on previously-wooded lands starting in 1970. Fourteen shallow observations wells were installed by U.S.G.S. and Suffolk County in the Spring of 197Z. The water table at that time was about forty feet below the land surface. C-3 F~GURE C-1. COMPARISON OF FIELD DATA WITH HALRAS SIMULATIONS OF NITROGEN UPTAKE BY TURF, 25' 20' 0 5 10 15 20 25 Rate of Nitrogen ( lb./1000 ft.2/32 months) , I STUDY AREA 0 WELL LOCATIONS RECHARGE BASINS ] RESIDENTIAL AREAS ] WOODED AREAS ] VACANT AREAS Twelve Pines FIGURE C-2. TWELVE PINES STUDY AREA, C-4 During the 197Z to 1978 period, data were collected to estimate the amount of nitrogen added to the soil. Since the area serviced by collective sewers which export sewage from the start, sewage was not a significant source. This allowed for closer analysis of the nitrogen sources associated with turf management, rainfall, and pets. Table C-Z summarizes the data used in the WALR. AS simulations. Turf simulations were combined with water and nitrogen simulations of forested land and impervious surface to calculate the overall nitrogen content of recharge water. The simulation results are summarized on Table C-3. rate. (1978). 1976 data. Pest waste nitrogen 16 kg/ha/yr clippings removed. Year Recharged (in) Recharged (lb/acre) C-5 Four of the observation wells were located in areas which received some recharge from outside of the study area, so data from these wells were not used. Data from the ten remaining wells (1,4,5,6,8~9,10,11,13~14) are summarized in Table C-4 for each year. Table C-4 Summary of Twelve Pines Well Data Total Nitrogen Wells 1,4,5~6,8~9~10~11~13~14 Number of Samples Mean (rog/l) Standard Deviation (mg/1) 197Z 40 1.58 1.14 1973' 8 1.05 0.64 1974 36 1.47 1.34 1975 40 2.17 1.49 1976 30 2.15 1.ZO 1977 ZO Z.19 1.36 1978 41 Z.55 1.03 * Samples not representative of the year. Data for May or June, 1 sample per well, only. The time it takes for recharge water to travel the 40 feet from the surface to the water table is approximately Z years (assuming general sand and gravel subsurface conditions). The simulated recharge nitrogen concentrations should thus be compared with nitrogen concentrations measured in wells two years later. Because flow in the unsaturated zone is variable, subsurface water mixes and the nitrogen concentrations observed in wells will actually represent a mixture of recharge water from several years. The simulation results are compared with the field measurements in Table C-5. The simulated concentrations correspond closely to the observed concentrations. Assessment of Uncertainty A sensitivity analysis was conducted on the turf simulations to determine the margin of error in the results, caused by the uncertainty of the accuracy of the way turf is represented by the model. The model was constructed by using the best estimated value for each parameter in order to accurately represent the system. For the sensitivity analysis the parameters were changed to represent the range of possible impacts that the system might have on nitrogen concentrations in recharge. The parameters were varied systematically to represent the maximum possible impact that the turf system would have on ground water nitrate levels and the minimum possible impact. Table C-0 shows the ranges of parameters that were used and the simulation results. C-6 Table C-5 Comparison of Observed and Simulated Nitrogen Concentrations in Ground Water, Twelve Pines Subdivision Year Wells Sampled Average Nitrogen Concentration in Wells (rog/l) Simulated Average Nitrogen in Recharge ( m gl 1) Year Root Zone Simulated (1974) 1.47 1.7 (1972) (1975) 2.17 Z.1 (1973) (1976) 2.15 2.7 (1974) (1977) 2.19 2.6 (1975) (1978) 2.55 2.0 (1976) average: Z.11 The maximum impact simulation resulted in a nitrogen concentration 10% greater than the medium impact; the nitrogen concentration of the minimum impact simulation was 46% less than the medium impact. The comparison of simulated nitrogen concentrations with observed concentrations on Table C-5 showed that in that ease the amount that the simulated concentrations differed from the observed concentrations was well within this range. These ranges were used in computing confidence intervals for land uses involving turf. 'FABLE {]-6. Maximum Impact Medium Impact Minimum Impact on Ground Water on Ground Water** on Ground Water o Depth of root zone: 10 cm Z0 cm 30 cm [3.9 in) (7,9 in] (11.8 in) o Parameter allowing increased N uptake via o Denitrification rate: 75 k~/ha 300 kg/ha 300 k~/ha (68.l lb/acre) (Z7Z.7 lb/acre) (g7Z.7 lb/acre) 0 0 0.5 0.0003 1/day 0.0005 1/day 0.001 1/day 0.0~$ 0.05 0.1 10.3 mg/l 9.4 rog/1 5.1 rog/1 D-1 Appe~&iz D Potato Nitrogen Simulations Leaching of agricultural fertilizers is a major concern on Long Island, and one of the objectives of this project has been to evaluate how various management practices can help to reduce these fertilizer losses. Since potatoes are the primary crop in this region, much of the research at the Long Island Horticultural Research Laboratory (L.LH.R.L) has centered around potato production. Wherever possible, we have used the results of such research in setting up and verifying our WALRAS potato simulations. Calibration of the Potato Model Between 197Z and 1975, research was carried out at the L.I.H.R.L on the effect of rate and timing of fertilizer nitrogen on yield and nitrogen recovery in Kat ahdin potatoes. Plots received 80 or 160 lb N/Ac either at planting or in split applications (Table D-l). Table D-1 Fertilization Rates and Timings for L.I.H.R.L. Experiments on Katahdin Potatoes, 197Z-1975 Treatment Fertilization during plant stage (lb N/A) Fertilization at planting 4-6" early (lb N/A) emergence high bloom Total quantity of nitrogen (lb N/A) 1 Z 3 4 6 80 0 0 0 80 40 Z0 Z0 0 80 40 0 20 Z0 80 160 0 0 0 160 80 40 40 0 160 0 0 40 40 160 D-2 Results from these field experiments were used for calibration of the WALRAS root zone model for nitrogen uptake by potato plants. Most fertilizer nitrogen is either taken up by plants or leached to ground water, with much lower losses to runoff and volatilization. Detailed field data are not available on nitrate leaching below potato fields, so data on plant uptake of nitrogen were used instead for the model calibration. Adjustments were made to the model parameters controlling uptake of nitrogen by the potato plants in order to make the model fit as closely as possible to the field data. Table D-Z compares the tuber nitrogen contents from these field experiments with those from our ~VALRAS simulations over the period from 197Z to 1975, and Figure D-1 shows a comparison of the results for these four years averaged together. On average over the four years, the WALI~AS results for tuber nitrogen were slightly low for the three treatments with 80 pounds of fertilizer nitrogen and slightly high for those with 160 pound~ of fertilizer nitrogen (Figure D-I). In all cases, however, the average simulated results were well within the range of field results for the four years measured. 140 120' 100' 80- 60- ~o. 2O KEY X = WALRAS results, average~ 1972 - 1975 0 = Field results, 2 3 4 5 6 Treatment Number FIGURE COMPARISON OF WALRAS SIMULATION RESULTS WITH FIELD MEASUREMENTS OF THE NITROGEN CONTENT OF POTATO TUBERS. THE FIELD EXPERIMENTS WERE CARRIED OUT BY THE LONG ISLAND HORTICULTURAL RESEARCH LABORATORY, D-3 Asse~ment of Uncertainty The WALRAS potato model includes many parameters in its simulation of plant uptake of nitrogen, and values for these parameters have been estimated from field data wherever possible. In order to determine how variability or uncertainty in these values affects the simulation results, we conducted a sensitivity analysis by systematically varying plant uptake parameters in the potato model. The results are shown in Figure D-Z. The error bars around each point in this figure show how the range of uncertainty in the plant uptake equations affects the predictions of nitrate concentrations in recharge water. The shaded points represent the conditions which best fit field observations, as previously discussed, and these values were used for the potato simulations in this report. 25 2C ,Jo do ,io ,~o 2~o Nitrogen Ferlilizer Applied (lb/acre) F!GURE D-2, SENSITIVITY ANALYSIS FOR THE WALRAS POTATO SIMULATIONS, THE ERROR BARS AROUND EACH POINT SHOW THE VARIATIONS IN SIMULATION RESULTS CAUSED BY VARYING PLANT UPTAKE PARAMETERS IN TIlE MODEL. D-4 Table D-~ Comparison of Field and Simulation (V/ALRAS) Results for Nitrogen Tuber N Content (lb Appendix E Glossary and Abbreviations Glossary Aquifer. A geological layer which is saturated and can yield significant quantities of water to wells and springs. Clustered Development. Grouping of houses into a higher than otherwise permissable density, leaving the remainder of the site in open space. Concentration. The weight of a substance that is dissolved in a given volume of water; milligrams per liter (rog/1 for short) is the unit used here, which refers to milligrams of the substance per liter of water. Contaminant. A substance which makes water less fit or unfit for human use or consumption and which is not present naturally. Evapotr~n-~piration. Processes which convert liquid water in the soil and on the surface into vapor; includes direct evaporation and transpiration by plants. Flow Path. The path along which ground water flows, usually under steady- state conditions. Ground water. Water saturating a geologic stratum beneath land surface; all water below the water table. Integrated Pest Management (IPM). The use of chemical, cultural, genetic, and biological pest control methods in ways designed to have minimal effect on nontarget organisms and the environment. Leaching. The process by which moving water removes a substance from a zone of soil; chemicals may be leached from the soil and carried downward by recharge water. Nitrate. A chemical compound containing one nitrogen atom and three oxygen atoms; nitrate in the diet reduces the ability of the blood to carry oxygen, especially in very young infants. Permeability. The property or capacity of a porous rock, sediment, or soil for transmitting a fluid; a measure of the relative ease of fluid flow. Recharge. The process by which water is added to ground-water storage, usually from the surface. Re~harge area. The location at which water can enter an aquifer directly or indirectly; generally an area consisting of a permeable soil zone and underlying rock material that allows precipitation or surface water to reach the water table. Root zone. The upper portion of the unsaturated zone, inside which plants and atmospheric processes influence water flow strongly and in which biological processes that decompose contaminants into basic chemicals (such as water and carbon dioxide) are the greatest. Saturated zone. The zone below the water table; all pore spaces are filled with water. E-2 Sim,,istion. Estimation of movement of water and dissolved chemicals to the ground water, based on site-specific calculations of processes including precipitation~ evaporation, plant uptake, runoff~ and leaching. Synthetic organic chemicals. Chemical compounds not normally found in nature that are synthesized by humans; include many plastics, herbicides, pesticides, and solvents; this group includes most of the worst toxic and hazardous materials to which people are exposed. Unsaturated zone. Zone between the land surface and the water table, inside which some of the pore spaces are filled with air rather than water. WALl{AS. The Water and Land Resource Analysis System, a system developed at Cornell University for simulation of movement of water and dissolved chemicals to and through the ground water. Water budget. An accounting of all of the inflows, outflows, and changes in storage of water within specific boundaries and over a specific time interval, such as a year. Water table. Top of the saturated zone. Abbreviations lb N/A/yr. Pounds of nitrogen per acre per year. mg/l. Milligrams per liter, referring to the number of milligrams of a substance per liter of water. A milligram is one thousandth of a gram. ug/1. Micrograms per liter. A microgram is one-thousandth of a milligram.