Environmental Science & Policy 5 (2002) 113–120
Fine tuning water quality regulations in agriculture to soil differences J. Bouma∗ , B.J. van Alphen, J.J. Stoorvogel Laboratory of Soil Science and Geology, Wageningen University, P.O. Box 37, 6700 AA Wageningen, The Netherlands
Abstract Groundwater quality has been defined in terms of threshold values for nitrate (50 mg l−1 ) and pesticides (0.1 g l−1 active substance). Variability in space and time, and cost and safety considerations have made it unattractive to verify water quality by repeated measurements. Proxy values have, therefore, been defined to characterise water quality. For nitrate, maximum allowable fertilisation rates have been specified and farmers have to apply the MINAS book-keeping system to keep track of their N-flows. For pesticides, listing of allowed pesticides functions as another proxy quality measure. Field tests and simulations on a Dutch farm demonstrated that water quality assessment using these proxy values does not correspond with direct assessment based on measurements and a comparison with the threshold values, which represent the true standard. A second problem is the generic character of the proxy methods, which do not reflect quite different nitrate and pesticide dynamics in different types of soil. These problems make the proxy approach quite problematic. We, therefore, propose the systematic introduction of information technology to be used for deriving soil-specific management practices that do not lead to an increase of the thresholds. Existing techniques for precision agriculture can be used, and the current registration of all parcels in The Netherlands in a geographical information system, including occurrence of different soil types, will be quite helpful. Such an information system on internet will allow better control than the current generic proxy systems and is likely to be quite motivating to farmers. © 2002 Published by Elsevier Science Ltd. Keywords: Groundwater quality; Environmental regulations; Soil quality; Environmental control systems
1. Introduction As long as food production was a major policy objective in post-war Europe, little attention was paid to the environmental side effects of agricultural production. Once food production started to exceed food consumption in the seventies of the twentieth century, the environmental impact of agricultural practices became a growing concern to groups of environmentally conscious citizens and politicians. Environmental protection laws were introduced, but their implementation proved to be slow, painful and difficult. Though farmers in western Europe currently constitute less than 5% of the population and their contribution to the gross national product is less than 5%, they still carry considerable political clout. This is partly due to the fact that their activities determine the state of the land in rural areas. Even in a densely populated country as The Netherlands, over 70% of the land is still used for agricultural production. Be that as it may, the question remains acute as to why the change to sustainable, economically viable farming systems is so difficult.
∗
Corresponding author. Tel.: +31-703-564-605; fax: +31-703-562-695. E-mail address:
[email protected] (J. Bouma).
1462-9011/02/$ – see front matter © 2002 Published by Elsevier Science Ltd. PII: S 1 4 6 2 - 9 0 1 1 ( 0 2 ) 0 0 0 3 0 - 8
The role of research in this overall process deserves special attention, as research should be in a prime position to provide the keys to innovative development. So far, agricultural research has played a key role in: (i) providing basic data for environmental laws and regulations, (ii) developing policy tools for implementation of environmental regulations, and (iii) developing innovative sustainable farming systems. This paper intends to analyse the role of research in the three above areas, and provides examples of how research can contribute to a successful implementation of environmental legislation. Attention focuses on the use of nitrogen fertilisers and pesticides, which currently receive ample attention in EU countries. Presented examples are based on research at a modern arable farm on prime agricultural land in The Netherlands.
2. The role of agricultural research 2.1. Basic data for environmental laws and regulations EU directives increasingly determine the threshold concentrations for nitrogen and pesticides in soil and groundwater, that are to be implemented at the national level. Threshold values reflect concentrations that are not harmful
114
J. Bouma et al. / Environmental Science & Policy 5 (2002) 113–120
to living organisms, including man, even when exposure occurs for an extended period of time. They are generally determined through ecotoxicological dose–effect experiments for a limited number of animal and/or plant species. A well-known threshold is the 50 mg l−1 nitrate concentration for drinking water (European Commission, 1991b), which also serves as the quality standard for shallow groundwater. Though often disputed for its weak scientific basis and the large assumed safety margin (Addiscott and Benjamin, 2000), it is the golden standard that has to be dealt with. Similarly for pesticides, a threshold concentration of 0.1 g active substances per litre is defined for water percolating below 1 m depth (European Commission, 1991a). This threshold is imposed uniformly, irrespective of the specific toxicity of different compounds. 2.2. Policy tools for implementing environmental regulations Well-documented threshold concentrations for agrochemicals are not sufficient for environmental regulation. Concentrations found under field conditions are bound to vary greatly in space and time. Aside from being quite costly, measurements are cumbersome as well, and protocols defining the number of samples to be taken and the necessary time intervals between samplings are often lacking (Droogers, 1998). To restrict operational difficulties, emphasis has shifted to the definition of proxy values. These can de determined more easily, but have only an indirect relation with the concentrations of chemicals in groundwater. For nitrogen fertilisation, the total amount of fertiliser that can be applied is used as a proxy value. For example, the EU nitrate directive of 1991 does not allow application of more than 170 kg N ha−1 in the form of organic manure (European Commission, 1991b). In The Netherlands, a farm-level accounting system for nutrients is currently being implemented, that registers net inputs (fertilisation) and exports (harvested product). The system, which is referred to as MINAS, defines limits for budgetary N surpluses (e.g. 100 kg N ha−1 for arable land in 2003) (Oenema et al., 1997) and imposes levies when these surpluses are exceeded. These surpluses correspond with the allowed (gaseous and leaching) losses. For pesticides, the Dutch ministry of agriculture published a multi-year crop protection plan (LNV, 1991) that listed 90 active ingredients destined for removal from the market by the year 2000. This listing acts as a de facto proxy value. Measured against the uniform principles for risk assessment applied in EU registration procedures (European Commission, 1991a), these chemicals are assumed to present unacceptable environmental risk. After considerable political commotion in the year 2000, involving a close 72 versus 73 votes in parliament, seven pesticides were classified as indispensable, and granted another 2 years on market. By 2003, the agrochemical industry should have developed environment friendly alternatives for these
compounds. Clearly, the definition of proxy values, such as maximum fertilisation rates and a listing of prohibited pesticides, contributes strongly to making environmental regulations operational. At the same time, the introduction of proxy values decreases the transparency of regulations, which can cause problems for their implementation. These problems are explored in this paper. 2.3. Innovative farming systems Application protocols for agrochemicals, such as nitrogen fertilisers and pesticides, have traditionally been derived by dose–effect experiments on small field plots or by laboratory experiments. Increasingly, however, the use of agrochemicals is considered in the much broader context of an entire farming system. Application of chemical or organic fertilisers does not only affect plant growth, but may also cause nitrate pollution in groundwater and/or ammonia emissions to the atmosphere. Also, the N concentration in plants may affect the occurrence of, and the damage caused by, pests and diseases. Pesticides may, similarly, cause pollution of soil and water and may also affect the quality of produce. By studying the entire production system rather than its separate parts, trade-offs can be made among contrasting demands. Trade-offs are part of the prototyping procedure that is currently used successfully to design innovative farming systems (Aarts et al., 2000; Vereijken, 1997). Important contributions to prototype analysis can be made by simulation modelling of nitrogen transformations and pesticide persistence in the soil environment (Hack-ten-Broeke et al., 1999; Boesten and van der Pas, 2000). Such analyses are indispensable tools in evaluating whether proxy methods, as discussed above, will succeed in keeping nitrate and pesticide concentrations below their respective environmental thresholds. Only use of calibrated and validated simulation models is feasible in this context because field experimentation with an adequate number of variants and repetitions is not feasible, for obvious economic and environmental reasons. This evaluation of proxy values is an important objective of this study, as we see that proxy values start to obtain a significance by themselves, while the link with concentrations of chemicals in groundwater and their relation to threshold concentrations defined in environmental laws tends to be forgotten.
3. Arable farming on a prime agricultural soil The implications of agricultural policies and their alternatives will be illustrated for a commercial, arable farm in The Netherlands. First, conventional management practices are analysed with reference to current proxy values for nitrogen fertilisation and pesticide use. Next, effects of alternative management practices, using the concepts of precision agriculture, will be analysed in the same manner. All studies were conducted in close co-operation with the farmers. Conventional and precision management are compared by
J. Bouma et al. / Environmental Science & Policy 5 (2002) 113–120
115
Fig. 1. Four management units in a Dutch arable farm as derived from a detailed soil survey according to the procedure of van Alphen and Stoorvogel (2000).
evaluating whether nitrate and pesticide concentrations in groundwater (at 1 m depth) are likely to remain below their environmental thresholds.
recommendations (Fr in kg N ha−1 ) are provided for individual fields, and are calculated as:
3.1. Study area
in which Nmin (kg N ha−1 ) is the average soil mineral N concentration measured at the start of the growing season. Similarly, recommendations for potato and sugar beet are calculated as:
The farm under consideration is commercially run and located in the central-western part of The Netherlands (51.7◦ N, 4.0◦ E). It covers an area of 85 ha and applies a crop rotation of winter wheat, consumption potatoes and sugar beet. Soils originate from marine deposits and are generally calcareous with textures ranging from sandy loam to clay. Within the farm, four management units are distinguished which were derived from a detailed soil survey according to the procedure described by van Alphen and Stoorvogel (2000) (Fig. 1). The soils are characterised as fine, mixed, mesic Typic Fluvaquents (Soil Survey Staff, 1998) or Mn25A–Mn45A on the Dutch 1:50,000 soil map (Vos, 1984). Soil variability is large and mainly expressed through differences in: (1) texture (the average clay content over 0–100 cm varies from 14 to 50%); (2) soil organic matter (SOM) contents (varying from 5 to 58 g kg−1 ); and (3) subsoil composition (peat or mineral soil). With excellent drainage conditions, controlled by a dense system of pipe-drains, the area is considered as prime agricultural land. 3.2. Conventional management 3.2.1. Nitrogen fertilisation Conventional fertiliser management is based on recommendations provided by the Dutch extension service. Their recommendations are founded on empirical dose–response relations that were established some 20 years ago. Over time, they have been adapted to meet the demands of new and more productive varieties. In the case of winter wheat, production is now targeted at 12 t ha−1 . Fertiliser
Fr,wheat = 300 − Nmin ,
Fr,potato = 245 − 1.1Nmin , Fr,sugarbeet = 200 − 1.7Nmin . The average Nmin concentration measured in different fields during the period 1997–1999 amounted to 60 kg N ha−1 . Based on this value, the average fertiliser recommendations for winter wheat, potato and sugar beet were 240, 180 and 100 kg N ha−1 . The allocation of land to the three major crops in rotation was on average 55 ha (64%) to winter wheat, 17 ha (20%) to potato and 14 ha (16%) to sugar beet. By combining these data, the average fertiliser input during the period 1997–1999 was calculated at 207 kg N ha−1 . Data on the average N uptake by crops were available for the period 1997–1999. Farmers have to calculate this uptake at the farm level and include it on the N balance for MINAS. Calculations are conducted by multiplying the average yield level for each crop by the average N content measured in samples of the harvested product. In the study area, the average N uptake over the period 1997–1999 was 157 kg N ha−1 . By subtracting the average crop uptake from the total fertiliser input (mineral and organic fertiliser) a net N loss can be calculated. According to MINAS guidelines, annual losses on arable land must not exceed 100 kg N ha−1 in 2003. If losses do exceed this value, a levee of US$ 2 (equivalent) will be charged for each kg of excess N.
116
J. Bouma et al. / Environmental Science & Policy 5 (2002) 113–120
Table 1 Farm level N balance for the period 1997–1999 1997 (kg N ha−1 )
1998 (kg N ha−1 )
1999 (kg N ha−1 )
Input Mineral N Organic N Total
221 41 262
248 60 308
204 80 284
Output Crop uptake Total
158 258
159 259
153 253
Balance Net N loss
104
149
131
Levy in 2003 (US$) Per hectare Farm level
10 850
123 10413
78 6588
N input consists of mineral and organic fertiliser, N output consists of crop uptake and net losses. Levies on N losses > 100 kg N ha−1 are fictive, as they will not be charged until 2003.
The farm level N balance for the period 1997–1999 is presented in Table 1. Considering these data, two issues attract attention. First of all, the MINAS guidelines for 2003 were clearly exceeded. As the future financial consequences are considerable (levies vary between US$ 850 and US$ 10,413 at the farm level), fertiliser management will have to be adapted to the new and stricter regulations. Obviously, this means a reduction of the total fertiliser input. A second issue concerns the steady increase of manure application. This effect is caused by a financial benefit: dairy farmers are paying arable farmers a US$ 2 kg−1 N bonus to apply excess manure on their land (Dutch dairy farmers generally produce more manure than they can apply on their own pasture). Based on the above, one may conclude that conventional N management will have to be adjusted before 2003 to meet the MINAS guidelines. Assuming that crop uptake will not increase dramatically, the maximum fertiliser input from 2003 and onwards will amount to 157 + 100 = 257 kg N ha−1 on average. As manure application provides financial benefit, the entire N quotum will be used annually. Most likely, mineral fertiliser will be applied following official recommendations (207 kg N ha−1 on average), leaving 50 kg N ha−1 to be applied in the form of manure. Compared to the situation described in Table 1, this means a reduction of fertiliser input by 28 kg N ha−1 on average. This is hardly a sharp decrease and it remains doubtful whether this reduction will suffice to keep or reduce NO3 concentrations in shallow groundwater below 50 mg l−1 . The proxies as defined by the MINAS system indicate that the farmer exceeded the 100 kg N ha−1 limit in 1998 and 1999, while the limit was met in 1997 (Table 1). The question that remains is whether this actually means that the threshold of 50 mg NO3 l−1 is passed. A mechanistic simulation model, called Water and Agrochemicals in soil and Vadose Environment (WAVE) (Vanclooster et al., 1994), was successfully calibrated for the local conditions on the farm
Fig. 2. NO3 concentrations in groundwater at 100 cm depth throughout 1997 for four management units in a Dutch arable farm under conventional management (winter wheat).
(van Alphen, 2002). The simulation model allows us to verify the nitrogen concentrations under different fields in the drainage water. In 1997, MINAS indicated that the nutrient balance of the farm was acceptable (104 kg N ha−1 surplus with 100 kg N ha−1 surplus being acceptable). The simulation results for NO3 concentrations in the leachate (Fig. 2) show that nitrate concentrations exceed the 50 mg l−1 threshold for NO3 in 1997. The increase was even greater in the other 2 years (not shown). The simulation results illustrate the limitations of the proxy approach at the farm level. Although the proxy, as defined in MINAS, is only slightly exceeded in 1997, the environmental threshold for groundwater quality, set by the European Commission, is clearly surpassed. 3.2.2. Pesticide use Pesticide use is controlled through registration procedures at EU and national level. During registration, it is determined whether a pesticide may be used, and under which specific conditions. Requirements applied for EU registration are described in the uniform principles for risk assessment (European Commission, 1991a). An important environmental criterion taken into account is leaching to the groundwater. Pesticide concentrations in percolating water must not exceed 0.1 g l−1 and risks of exceeding this threshold are primarily assessed through simulations for a number of standard scenarios. A scenario, in this respect, refers to a characteristic combination of soil and weather conditions. At the EU level, nine scenarios are evaluated, that together represent a variety of soil and weather conditions encountered in Europe (FOCUS, 2000). If all environmental criteria are satisfied for at least one scenario, an active ingredient may be granted access to the European market. National registration procedures, however, have the final word in determining whether a chemical may actually be used in a specific country (Rasmussen and MacLellan, 2001). Often this includes a second evaluation of pesticide persistence in soil and risks for percolation to the groundwater. Again, simulation studies are conducted for a varying number of characteristic scenarios (e.g. two scenarios are used in
J. Bouma et al. / Environmental Science & Policy 5 (2002) 113–120
117
Table 2 Pesticides used in the study area during the period 1995–1999 Pesticide
Active ingredient(s)
Typea
Cropsb
DT50 (days)
Agil 100 Ec Allegro
Propaquizafop Epoxiconazole Kresoxim-methyl Metsulfuron-methyl Bentazon Cymoxanil Mancozeb Isoproturon Glyfosfate Metribuzin Fluazinam Fluroxypyr Chlorothalonil Propamocarb Rimsulfuron Clodinafop-propargyl Ethofumesate Bifenox Mecoprop-P
H F
P, S W
H H F
W P P
H H H F H F
W P, W P P W P
H H H H
P W S W
10 280 0.6 31 48 0.7 5 30 8 34 107 27 10 25 31 0.6 37 8 11
Ally Basagran Curzate M Isoproturon Round Up Sencor WG Shirlan Starane Tattoo C Titus Topik 240 Ec Tramat Verigal D
Kom (l kg−1 ) 242 940 120 28 0.4 10 1000 54 6533 32 3225 35 5031 179 35 816 84 1420 <1
Application rate (kg ha−1 ) 0.10 0.12 0.12 0.03 0.96 0.09 1.36 1.00 0.72 0.30 0.15 0.14 1.00 1.00 0.01 0.05 0.30 0.62 0.75
Half-lives (DT50) and sorption coefficients (Kom ) were derived from Linders et al. (1994). Recommended application rates are specified after Oomen et al. (1999). a H = herbicide, F = fungicide. b P = potato, S = sugar beet, W = wheat.
Denmark, whereas a single scenario is used in The Netherlands and Germany). Lysimeter experiments may complement simulation results if excessive leaching is suspected. Pesticides used in the study area are listed in Table 2. Commercial names were provided by the farmer, chemical names (active ingredients) were derived from Oomen et al. (1999). All chemicals have passed EU and national registration procedures and are applied following official recommendations in the Dutch crop protection guide (Oomen et al., 1999). A key question, however, is whether international and national registration procedures can guarantee that pesticide use is environmentally safe at the farm level? Recent studies by van Alphen and Stoorvogel (2002) conclude that this is very doubtful. A major problem lies in extrapolating the results of scenario studies, which only consider a very limited number of “representative” soil profiles, to larger areas. Extrapolation is particularly difficult, since pesticide persistence in soil has been shown to present great spatial and temporal variation (Walker et al., 2000). Therefore, registration based on standard scenarios can result in large over- or under-estimation of the risk associated with pesticide use in a specific region or on a specific farm. van Alphen and Stoorvogel (2002) carried out a relative risk assessment for 19 pesticides from which three risk-bearing pesticides: isoproturon, metribuzin and bentazon, were selected for more detailed study. The results are presented in Fig. 3. Although all three pesticides are approved for use, only the use of isoproturon carries minimal risk, whereas the use of bentazon is associated with a high risk to exceed the threshold value of 0.1 g l−1 active
substance. Again, it is shown that the proxies do not guarantee that concentrations do not exceed thresholds. 3.3. Alternative management strategies The central objective is to reach the thresholds set for groundwater quality. One procedure that can be followed is to define the proper proxies. However, one can argue whether this can be done by simply developing the proper regulations without the development of alternative management strategies that allow farmers to fulfil the criteria. Groundwater quality is based on highly variable (both temporal, as well as spatial) processes. Up till now, proxies for nitrogen fertilisation and pesticides mitigate this variability. What does this mean in the real world? Farmers are constrained in their fertilisation in areas where the risks for high nitrogen leaching are minimal. On the other hand, in other areas, farmers are allowed to apply certain quantities of fertiliser that lead to the contamination of groundwater above the thresholds. On the case of pesticides, we see that pesticides are banned from the market, whereas they can be used without risk in large areas. The solution can be found in the development of alternative management techniques that deal with the spatial and temporal variability in growing conditions. Precision agriculture has been promoted as an alternative management strategy that allows farm management to properly manage this variability (Robert et al., 2000). As a result, agrochemicals are used more efficiently and its use can be reduced leading to a reduction of costs for the farmer, and
118
J. Bouma et al. / Environmental Science & Policy 5 (2002) 113–120
Fig. 3. Simulated leaching of three pesticides in a Dutch arable farm.
at the same time a reduction of the environmental impact. Let us look in details as to what this means for nitrogen fertilisation and pesticide use. 3.3.1. Nitrogen fertilisation For the farm, a simulation model and real-time weather were used to monitor soil mineral N levels in a number of soil units (van Alphen and Stoorvogel, 2001). Early warning was provided when soil mineral N concentrations dropped below a critical threshold. Used as a trigger, this information served to optimise the timing of four consecutive N fertilisations. Fertiliser rates were determined through exploratory simulations that calculated the amount of mineral N required under normal weather conditions. The application of this procedure resulted in a reduced fertiliser input by 14–32% (ranging from a reduction of 31 kg N (management unit 1) to 71 kg N (management unit 4). The reduction was achieved without affecting grain yield. Considering both the timing and the specific local soil conditions resulted in efficiency gains attained through precision N management. The results, in term of the corresponding NO3 concentrations, are shown in Fig. 4 for the four management units. Under this alternative management practice, the NO3 concentrations stay below the 50 mg NO3 l−1 . 3.3.2. Pesticide use In terms of environmental risks, differences in pesticide leaching within and among fields were apparent (Fig. 3). Bentazon showed a significant risk to exceed the threshold value, but has been banned by the farmer, as good alternatives have become available. Metribuzin is probably the most interesting case, since a few fields show a high risk for exceeding the threshold, but in several other fields metribuzin can be used with little risk.
In terms of precision management, the results show that bentazon should be banned for the entire farm. However, the farmer can apply metribuzin on several fields with little risk but has to find less persistent alternatives for those fields representing a high risk. 3.4. Implications for policy Environmental indicators for groundwater quality, defined in terms of threshold nitrate and pesticide contents, have been established for the European Union and are broadly accepted. The very high spatial and temporal variability of these parameters, as well as budgetary and safety considerations have resulted in definition of proxy values for both nitrate and pesticide pollution of groundwater. These proxy values, as discussed in this paper, can be monitored and can
Fig. 4. NO3 concentrations in groundwater at 100 cm depth throughout 1997 for four management units in a Dutch arable farm under precision management (winter wheat).
J. Bouma et al. / Environmental Science & Policy 5 (2002) 113–120
therefore be part of regulatory protocols. This study has, however, demonstrated that conditions indicated by proxy values can be within the limits of the regulations, while the corresponding concentrations in groundwater are well above threshold values specified in environmental laws. This is unacceptable and illustrates the need to test any proxy value being proposed in terms of its corresponding nitrate and pesticide contents in groundwater. For this purpose, calibrated and validated simulation models can be used as illustrated in this study. The soils that occur on the farm being studied here qualify as prime agricultural land and belong to the best agricultural land in The Netherlands. If proxies do not adequately reflect environmental thresholds in these highly adsorptive soils, deviations are likely to be more severe in lighter textured soils. However, not only the proxy system is problematic, the lack of distinction of different soil types also presents serious problems. This is being recognised by the regulatory agencies in The Netherlands. In the Staatscourant (official government paper, in which all new laws are presented) of 6 July 2001 (no. 128:11), new proxy values are presented for nitrogen fertilisation in sand and loess soils. Again, no link is made here with nitrate contents of groundwater. This appears to be an ad hoc unsatisfactory course of events. Why add these two soil types and no others? The question must be raised, whether the time has not come to revise the overall regulatory process making use of modern information technology. This is particularly relevant because the same government paper mentions progress with the “registration of parcels of land” which will be completed in 2002. In this program, all farmer’s fields in The Netherlands will be represented in a geographical information system on internet, including data on land use, fertility status and management practices. Most farmers have internet, and this new arrangement presents an excellent opportunity to remove the generic and quite abstract character of current regulations which does not create any affinity with farmers. When, in contrast, a farmer deals with his own fields and uses the information system to document his management practices and their effects, he is likely to be very interested and committed. Besides, such a system will drastically improve the transparency of the overall regulatory procedures providing clear-cut control mechanisms. This is very important, because the current MINAS system is very difficult to run. Recently, the official governmental agency for quality Control (De Algemene Rekenkamer) has written a highly critical report about the implementation of the MINAS system, which is way behind schedule. It is, in fact, simply impossible to run a top-down control system for some 100,000 farmers, that is based on highly detailed paper forms that have to be sent to a central office to be processed. It is time now to move ahead and try to implement a system based on modern information technology which takes into account natural soil variability. This fits in perfectly well with precision agriculture as a means to fine tune
119
farm management to the needs of the plants, thus creating substantial savings as was demonstrated in a farm study by van Alphen (2002). The shift to a modern information technology-based system has major implications for soil science. Soil surveys on farm level will be needed, as described by van Alphen (2002). We have found that this does not present any problems to the farmer. The investment is relatively minor (US$ 10,000 equivalent for a 100 ha farm), and when the survey is used in the context of the “parcel registration project” mentioned above, the farmer will have an excellent new management tool available. This is also important because modern concerns about food quality will require ever better documentation about what happened in the entire food-chain during production (“tracking and tracting”). Precision agriculture provides the ideal tools for this purpose. Soil science will have to generate data on nitrogen regimes and pesticide adsorption for the 1500 soil units in The Netherlands. Running models on real-time basis was used by van Alphen (2002) when assisting a farmer in implementing precision agriculture in practice. All this will require more attention to the logistics involved, but could provide a welcome fresh impulse to environmental regulations and their acceptance by farmers. References Aarts, H.F.M., Grashoff, C., van Keulen, H., 2000. Managing Dairy Farming Systems for Groundwater Conservation in the Sandy Regions of The Netherlands. Report 101, AB-DLO, Wageningen. Addiscott, T.M., Benjamin, N., 2000. Are you taking your nitrate? Food Sci. Technol. Today 14 (2), 59–61. Boesten, J.J.T.I., van der Pas, L.J.T., 2000. Movement of water, bromide and the pesticides ethoprophos and bentazon in a sandy soil: the Vredepeel data set. Agric. Water Manage. 44, 21–42. Droogers, P., 1998. Time aggregation of nitrogen leaching in relation to critical threshold values. J. Contam. Hydrol. 30, 363–373. European Commission, 1991a. Council Directive 91/414/EEC of 15 July 1991 concerning the placing of plant protection products on the market. Off. J. Eur. Commun. L230 (8), 1–32. European Commission, 1991b. Council Directive 91/676/EEC of 12 December 1991 concerning the protection of waters against pollution caused by nitrates from agricultural sources. Off. J. Eur. Commun. L375 (12), 1–8. FOCUS, 2000. FOCUS Groundwater Scenarios in the EU review of active substances. Report of the FOCUS Groundwater Scenarios Workgroup, EC Document Reference Sanco/321/2000 rev.2. Hack-ten-Broeke, M.J.D., Schut, A.G.T., Bouma, J., 1999. Effects on nitrate leaching and yield potential of implementing newly developed sustainable land use systems for dairy farming on sandy soils in The Netherlands. Geoderma 91, 217–235. Linders, J.B.H.J., Jansma, J.W., Mensink, B.J.W.G., 1994. Pesticides: benefaction or Pandora’s box? A synopsis of the environmental aspects of 243 pesticides. Report no. 679101014, RIVM, Bilthoven, The Netherlands. LNV, 1991. Meerjarenplan gewasbescherming. The Hague, The Netherlands. Oenema, O., Boers, P.C.M., van Eerdt, M.M., Fraters, B., Meer, H.G., van der Roets, C.W.J., Schröder, J.J., Willems, W.J., 1997. The nitrate problem and nitrate policy in The Netherlands. Report 88, Research
120
J. Bouma et al. / Environmental Science & Policy 5 (2002) 113–120
Institute for Agrobiology and Soil Fertility (AB-DLO), Haren, The Netherlands. Oomen, P.A., Marsman, H., Oostelbos, P.F.J., Schoenman-Weerdesteijn, M.E., Wanningen, R., 1999. Gewasbeschermingsgids, Vol. 15. Plantenziektekundige Dienst, Wageningen, The Netherlands. Rasmussen, K., MacLellan, M.A., 2001. The control of active substances used in biocides in the European Union by means of a review regulation. Environ. Sci. Policy 4, 137–146. Robert, P.C., Rust, R.H., Larson, W.E. (Ed.), 2000. Proceedings of the Fifth International Conference on Precision Agriculture, Bloomington, MN, 16–19 July 2000. ASA, CSSA, SSSA, Madison, WI. Soil Survey Staff, 1998. Keys to Soil Taxonomy, 8th Edition. USDA-NRCS, Washington, DC, USA. van Alphen, B.J., 2002. A case study on precision nitrogen management in Dutch arable farming. Nutrient Cycling in Agro-ecosystems, in press. van Alphen, B.J., Stoorvogel, J.J., 2000. A functional approach to soil characterisation in support to precision agriculture. Soil Sci. Soc. Am. J. 64, 1706–1713.
van Alphen, B.J., Stoorvogel, J.J., 2001. A methodology for precision nitrogen fertilisation in high-input farming systems. Precision Agric. 2, 319–332. van Alphen, B.J., Stoorvogel, J.J., 2002. Effects of soil variability and weather conditions on pesticide leaching—a farm level evaluation. J. Environ. Quality, in press. Vanclooster, M., Viane, P., Diels, J., Christiaens, K., 1994. WAVE, a mathematical model for simulating Water and Agrochemicals in soil and Vadose Environment. Reference and users manual, Institute for Land and Water Management, Leuvem, Belgium. Vereijken, P., 1997. A methodological way of prototyping integrated and ecological farming systems (I/EAFS) in interaction with pilot farms. Eur. J. Agron. 7, 235–250. Vos, G.A., 1984. Bodemkaart van Nederland: toelichting bij kaartbald 37—west Rotterdam. STIBOKA, PUDOC, Wageningen, The Netherlands. Walker, A., Jurado-Exposito, M., Bending, G.D., Smith, V.J.R., 2000. Spatial variability in the degradation rate of isoproturon in soil. Environ. Pollut. 111, 407–415.