Previous Page Table of Contents Next Page

Chapter 5: Summary and recommendations

Specific actions recommended for the control of sediment, fertilizer and pesticide impacts on water quality have been enumerated in each respective chapter. In this concluding chapter are discussed a selection of issues pertaining to water quality which have over-arching implications for agriculture. Some, such as internalization of all environmental costs, represent a significant challenge to the way in which agricultural agencies pursue agricultural practices. Others reflect the interaction between agriculture and other national development priorities and reflect public policy issues that are of national significance.

In his analysis of impacts of agriculture on water quality in the Brazilian state of Paraná in which agricultural expansion has had major impacts on water quality, Andreoli (1993) summarized a series of institutional and substantive lessons from that state into the following observations that have general application.

· The environmental impacts on water resources [in general] caused by agricultural activities cannot be disassociated from the agricultural impacts in production areas themselves. They require monitoring, and preventive measures should always be systemically integrated.

· It is necessary to develop and implement water resource monitoring systems with a prior definition of indicators, parameters, tolerance limits, frequency and sampling points, combining this information with quantity data.

· Data and information generated should be properly treated in the sense of disseminating them as much as possible in order to heighten awareness and mobilization of the public sector and of society with respect to agriculture's impact on the environment.

· Attempts should be made to exchange information and to pursue horizontal cooperation among countries, in order to promote the exchange of information and experiences.

· In the prevention systems proposed, solutions to causes should be looked for, seeking to match the agricultural model to the socio-economic needs of the population within environmental limits and vocations.

· Besides treating water quality related problems, there is evidence of other problems generated by conflicts in use, particularly the need to integrate quality management with the quantity of water within a comprehensive, decentralized and participatory management system, reconciling regional development with environmental protection.

· The cooperation of fiscalizing and monitoring agencies dealing with agrotoxic substances is urgent, with the capacity for control structures, seeking the development of biological indicators (enzymes, AMES test, biotesting, bioindicators) of residues and of anatomy and pathological damages caused by agrotoxic substances.

Necessity to internalize costs at the farm level

Ultimately, any strategy to reduce agricultural impacts on water quality will only be successful if it is implemented at the farm level. Therefore, implementation of control measures at the farm level will only be successful and sustainable if the farmer can determine that it is in his economic interest to undertake such measures. Therefore, the economic benefits from such factors as implementation of erosion control measures to maintain soil fertility, capital costs associated with improved manure handling and distribution, etc., must be clearly seen to be offset by reduced energy consumption in minimum till situations, improvement in soil fertility by improved manure handling and erosion control, reduced fertilizer costs, etc.. This implies that agricultural agencies must use a holistic approach to the economics of farming practices. There are abundant examples from both developed and developing countries that indicate that this approach is equally applicable to all farmers who have a long-term interest in their land.

In cases where particularly serious pollution of surface and/or groundwater (such as creation of a groundwater recharge reserve) creates conflicts over water rights and beneficial uses, mitigation is often addressed by a mixture of regulatory and voluntary measures. These measures may involve change in agricultural use or land management practice, or may take land entirely out of production. Where the cost-benefit is not in the farmer's favour, compensation becomes an important issue. While compensation is a well established legal recourse in developed countries, appropriate compensation for land owners in cash or kind should be considered as part of pollution mitigation programmes in developing countries. The situation in the former Soviet republics is particularly unique; the conversion of public agricultural land to private ownership requires new administrative, compensatory, and infrastructure costs to achieve water quality protection. In some countries, the costs of social disruption of landless farm labourers needs to be factored into the overall compensatory package.

Integrated national water quality management

The need for integrated water resources management has been widely accepted as a necessary national policy goal (ICWE, 1992; United Nations, 1992, World Bank, 1993; FAO 1994c). From the agriculturist's perspective, only an integrated approach permits the evaluation of the role of agriculture in a national water resource management programme, and protects against disjointed, inefficient and inequitable policy decisions for water quality remediation. The disastrous environmental situation in many Eastern European countries as well as in some rapidly industrializing countries, provides ample evidence for the types of policy actions that need to be taken to deal cost-effectively with the role of agriculture within the larger framework of water quality management.

The example of Lithuania is instructive (FAO, 1994b). The following seventeen issues form the basis for the water quality policy that is proposed in Lithuania. Clearly, agriculture is but one of the concerns that comprise the mix of water quality policy issues. Also, the emphasis on piggeries can be replaced by other types of intensive animal farming in other countries.

1. Unsafe groundwater supplies in rural areas.

2. Need for identification/definitions of water protection areas.

3. Waste from pig complexes is discharged directly into watercourses; lack of appropriate technology for handling of waste from pig complexes.

4. Conflicting views on (a) intensive agriculture, and (b) water quality protection from nutrient and chemical pollution.

5. Groundwater pollution from municipal waste water in Karst areas.

6. Policy issues related to the use of surface water are linked with pollution and dilution requirements during periods of low flow.

7. Conflicting functions of (a) licensing and (b) quality control of pesticides, carried out by the same agency.

8. Financing and implementation of municipal waste treatment to EC-HELCOM (Helsinki Commission) recommended standards.

9. Collection/treatment of industrial/urban rainwater runoff.

10. Poorly managed solid waste dumping sites.

11. Consideration of step-wise, time-targeted implementation of industrial effluent standards.

12. Need to consider economic and social consequences of effluent standards and water quality objectives.

13. Industry to have access to clean technology for safe disposal of waste.

14. Inefficient and operational problems of combined industrial and municipal waste water treatment.

15. Viability of irrigation is questionable; alternative methods for disposal of slurry waste from pig complexes have to be identified. [Note: refers to use of irrigation to dispose of slurry waste from pig complexes.]

16. Water protection and general environmental health policies are not consistent and need to be adapted to general policies of extended private initiative and responsibility.

17. Responsibility, ownership and mechanisms for certification, accreditation and control of water quality laboratories including main objectives (environmental, hygienic, law enforcement and scientific purposes) need to be defined.

Missing from this list is improved pest management and fertilizer management, particularly from the context of efficient use of animal wastes and the environmental (including water quality) and public health benefits from a more optimal use of pesticides. In other countries with different climatic and topographic characteristics, additional policy elements would include:

· erosion control

· aquaculture impacts

· deforestation impacts

· wetlands management

· pesticide management and control (importation, manufacture, sales and application, and disposal)

· irrigation management

· rice/fish culture management

Assessment methodology

In Chapter 1 it was noted that a typical dilemma in managing water quality in river basins was the difficulty in determining the extent to which agriculture contributes to the overall water quality problem. Therefore, recommendations for assessment of impacts of agriculture on water quality fall into two spatial categories - one is at the farm level, the other is at the river basin level. Assessment and decision-making at these two scales are fundamentally different.

At the farm level, assessment and decisions are those that can be implemented by the farmer. While the net benefits may be felt at larger scales, the objective is to implement decisions that are practical and which are economically advantageous for the farmer. At the river basin scale the user is a regional or national agency which must assess the contribution of agriculture to the larger problem of river pollution, and the management alternatives both for point and non-point sources that could be implemented in different parts of the basin that would have the greatest impact at the least cost. Therefore, basin-scale assessment is essential for the development of rational and cost effective remediation and control programmes which are usually driven by national policies on water pollution. The tools for agricultural impact assessment are well known, however, they need to be systematized into a general methodology and integrated with modern environmental chemistry and take advantage of new advances in the field of information technology. The methodology needs to be developed at two levels of detail: (a) at the screening level in which a rapid assessment can be made and which provides approximate levels of predicted impacts based on easily accessible information; and (b) at a detailed level for use in detailed studies for the purpose of developing remediation options.

At the small scale (field, small catchment) it is important to be able to predict the nature of agricultural runoff and associated loss of nutrients, sediments and pesticides, so that agriculturalists can predict the probable impact of different crop and land management options. Table 13 identifies a selection of models that are used for agricultural assessment purposes. It is recommended that one or more screening models be adapted for use in developing countries for the purpose of estimating erosion and chemical loss at the plot or field level, and for "gaming" with alternative land use options relative to erosion and potential for chemical runoff. Appropriate screening models should not require excessive data, should integrate over seasons, and be easily transportable from one region to another. Such models will require selective calibration by national agricultural agencies to ensure reliability.

At the large scale (sub-basin, basin, regional) there is an urgent need to develop simple methods for making estimates of erosion at the basin scale, as has been discussed more fully in Chapter 2. Also at the basin scale there is a need to develop a systematic methodology that permits evaluation of water quality impacts of agriculture relative to other types of polluting sources. One possibility is a combination of the "rapid assessment" methodology of the World Health Organization (Economopoulos, 1993) in which point sources can be rapidly inventoried and assigned to pollutant categories together with a non-point source assessment methodology (a screening type of model), both operated within a georeferenced expert system.

Environmental capacity

GESAMP (1986) defines environmental (also known as receiving, absorptive or assimilative) capacity as "the ability of a receiving system or ecosystem to cope with certain concentrations or levels of waste discharges without suffering any significant deleterious effects" (Cairns, 1977, 1989). All activities, including agriculture have some level of impact on water quality; the issue is whether the impact exceeds thresholds which society deems unacceptable for social, economic or cultural reasons. Scientifically, much is already known - rates of uptake, dilution, losses through volatilization, etc. However, determination of the threshold values requires not only scientific input, but also socio-economic and cultural inputs. According to GESAMP, this works well within an interactive environmental management strategy (Barg, 1992).

For agriculture, there is need to determine what the environmental capacity is for different types of runoff products in the local context. Determination of acceptable threshold values could be systematized through use of an expert system in which the database includes all relevant scientific characteristics; the decision on the threshold value are evaluated as a series of options that are filtered through the knowledge (input) of local circumstances.

The data problem in water quality

(a) Data Programmes: It was noted in Chapter 1 that, typically, the data which are needed for such an assessment are not often available from conventional water quality monitoring programmes. However, there are new techniques in environmental chemistry and toxicology which permit much more cost-effective determination of the nature and source of water quality problems than is possible with conventional physico-chemical measurements that are carried out in most water quality programmes. Applegren (FAO, 1994b), in his assessment of water resource policy needs in Lithuania, a country which is probably typical of many other former Soviet republics, identified the need for strengthened water quality monitoring and a water resources database.

The framework for cost-effective improvements in water quality measurement is as follows:

· Reduce fixed site monitoring networks and expand the survey approach to water quality measurement (Rickert, 1993).

· Achieve a better balance between water sampling (traditional) and other media such as suspended sediment sampling. Note that hydrophobic contaminants such as organochlorine compounds including many older pesticides, PAHs and PCBs are often found only at trace levels in water but can be easily determined on solids. It follows that management judgements about presence/absence/toxicity and environmental and/or human health impacts of aquatic chemistry are frequently grossly in error if analysis is made only on water samples.

· Use of "Environmental Effects Monitoring" (EEM) to reduce the amount of analytical chemistry and to increase environmentally relevant information for decision-making. Commonly, EEM involves in-stream assessments using biological survey techniques. The rationale is based on the fact that, if aquatic contamination exists, it will be evident in biota (Reynoldson and Metcalfe-Smith, 1992; Reynoldson et al., 1995; US-EPA, 1989). Canada and the United States have agreed on the use of biologically-based objectives for the management of water quality of the Great Lakes. These techniques are very useful in developing countries insofar as the measurements require training in biology (usually good in developing countries) and is labour intensive rather than capital intensive. EEM reduces reliance on high-end analytical chemistry with its large capital costs and lengthy training. EEM can involve many other types of biological measures such as fish health, fecundity, immune suppression, etc. Many of these tests are simple and are low cost.

· Except where specific chemical measures are justified, make use of modern screening techniques that provide economical indicators of chemical presence and/or impact. Screening techniques are designed to provide quick information which guides the decision on where more detailed (and expensive) chemistry is needed. Screening techniques eliminate the "menu" approach to environmental chemistry wherein laboratory analysis is restricted, for reasons of cost, to a predetermined set of chemical parameters. Experience has shown that such lists frequently have little value in studies of environmental contamination. Screening tools include the following types of activities:

· Standardized laboratory bioassays (Keddy et al., 1994)

· Immunoassay tests (Bushway et al., 1988; Thurman et al., 1990)

· Measures of fish health such as red/white blood cell ratio, slime on body and gills, etc.

· Use of enzyme measures to determine whether fish have been exposed to toxic chemicals (e.g. MFO induction [Mixed Function Oxidases])

· Measure of total chlorine (an indicator of total chlorinated compounds).

· TIE (Toxicity Identification Evaluation).

(b) Mobilizing Data: It is essential that agriculturalists can determine the impacts of agriculture on water quality at the medium and large scales relative to other potential sources of water quality degradation. This requires that conventional monitoring data for nutrients, salinity, and suspended sediment need to be integrated into a single information system that permits analysis of basin or sub-basin trends in water quality and sediment transport relative to point and non-point sources and gross indicators of land use, topography, soils and climate. That this is not done in most countries reflects inter-agency and institutional problems and the lack of suitable software. Measures of data reliability must be established and confidence levels established for interpretive activities. Analysis of existing data rapidly leads to identification of data gaps, of unreliable parameters, and of parameters which have no useful function and which can be eliminated from monitoring programmes.

Water quality indices for application to agricultural water quality issues

Water Quality Indices refer to two or more parameters that indicate the "healthiness" of water. In some cases, indices reflect ecosystem behaviour; in other cases, they indicate conditions of the aquatic environment (e.g. toxicity). These indices are generally designed to determine potential for ecosystem dysfunction, and to provide insight into pollution sources and management decisions for source control. Indices tend to be mainly used for descriptive purposes and are not often used directly for water quality management at the field level. The need is for indices that permit rapid assessment of impact of agricultural runoff and which can be used to make judgements on levels of impacts in space and time as a basis for management decisions concerning the need for controls. There is also potential to develop indices which link water quality impacts to economic factors relative both to upstream sources and downstream consequences, as a way to evaluate the economic impacts of agricultural runoff.

Currently, water quality indices are of the following types:

(a) Numerical Indices based on conventional water chemistry: There are commonly a half dozen indices that combine various chemical measures of water quality into an integrated index. These commonly include a mixture of nutrients, microbiology, dissolved oxygen and, occasionally, metals. Generally, these are used as descriptive tools for assessing river reaches. The more successful indices use a limited number of parameters (e.g. a eutrophication index would use nutrients and dissolved oxygen or BOD) and describe one type of water pollutant impact (e.g. eutrophication). The most complete recent reference is a report (in Dutch) produced for RIZA (Netherlands) on chemical indices (RIZA, 1994).

(b) Effects Indicators/Indices: There are a wide variety of "effects" indicators which are often combined into an index. These generally are some measure of biological reaction to aquatic pollutants. Many of these are used as "screening" tools (as noted above) which can help managers resolve spatially and causally, the nature and intensity of the pollutant impact. These indicators include:

Bioassay: Usually a scoring system based on the performance of a number of standardized lab assays using pollutant-sensitive species that are indicative of various trophic levels (e.g. bacteria, algae, invertebrates, vertebrates). Bioassay tends to focus on toxicity impacts (Keddy et al., 1994).

Biotic Indices: There are a variety of standardized biotic indices that are commonly used in Europe for water quality assessment and management. Generally, these indices are developed from benthic assemblages in rivers and streams. The index represents the nature of benthic response, mainly to organic pollution (domestic and municipal wastes). These have not been very successful for toxics assessment. There have been many reviews of biotic indices (Reynoldson and Metcalfe-Smith, 1992; Metcalfe-Smith, 1994).

Ecosystem indicators: Increasingly, there is interest in indicators that describe how parts of ecosystems respond to physical and chemical stress. Indicators can include a range of ecological measurements (fish, benthic organisms, habitat, etc.). This technique has proven useful as a means of establishing norms against which managers can measure success of remediation measures (US-EPA, 1989). This approach has also been used in the United Kingdom, Canada, and Australia (Reynoldson et al. 1995)

Other Effects Indices: There are a wide range of indices that are used to assess nutrient and/or toxic stress. Many of these use fish as a useful surrogate for impacts on humans. These indices include:

- measures of fish health using histological (e.g. red/white cell ratio) and pathological measures (size and appearance of organs).

- presence/absence of contaminant metabolites in fish bile, liver, etc.

- presence of enzymes as part of detoxification process in organisms (e.g. measures exposure of fish to toxic chemicals).

(c) Other Chemical Indicators: These are integrating (and usually simplifying) chemical measures of groups of compounds. An example is the mapping of chlorine residuals as a measure of total chlorinated material in the water column in a river basin. "Hot spots" within the basin indicate potential problem areas. The objective of such indicators is to use simple, inexpensive measures to determine whether problems may exist and to guide decisions on priorities for further (and expensive) chemical analysis.

General Note: Many of these techniques are lower in capital equipment requirements and higher in useful information than conventional water chemistry, and are capable of implementation by developing countries which typically have good capacity for biology but much less for advanced environmental chemistry. Such indices do require a shift in the "data paradigm" which continues to be dominated by the (western) chemical approach to water quality assessment.

Economic analysis of cost of water pollution attributed to agriculture

Because agricultural water pollution is of a non-point source nature, the quantification of pollutants and their impacts is more difficult than for point sources. However, the world's ever-increasing demand for dwindling supplies of good-quality freshwater requires that countries adopt a holistic approach to water resource management. Pollution control is now so expensive that decisions on resource management priorities should be guided by knowledge of the cost of water pollution to the various economic sectors. That cost is in two parts: the first is the direct cost (e.g. treatment) of meeting minimum water quality standards required for various uses; the second is the cost of lost economic opportunity because of inadequate water quality. Examples include: reduced production due to excessive salinity in irrigation water, and loss of fish production due to reproductive and growth impairment caused by toxic chemicals. It is only by knowing both direct and indirect costs, and by assigning these costs to the various economic sectors (including agriculture) that the true cost both caused by and absorbed by agriculture, can be evaluated relative to other sectors.

Information technology and decision making

An often cited benefit of information technology is the ability to electronically access data, text, graphics, etc. from an infinite number of locations in the world. An example for pesticide information was provided in Chapter 4. The hardware and software (e.g. World Wide Web on the Internet) is now reliably dedicated to such tasks to the point where information overload is now a problem. Nevertheless, while this type of information technology is the best known aspect of the information revolution, it is only one side of the information technology equation. The second side deals with the problems created by this ease of access. This includes the frequent absence of quality control and other meta-data which are needed to describe the characteristics of the data/information, and the immense problem of what to do with such large amounts of information when it is received and how to use it for decision-making purposes. Indeed, the challenge is no longer that of accessing information but one of integrating information in a systematic manner for the purpose of making management judgements about particular projects and problems in agriculture in general, and the management of water quality in particular.

Information technology is now conventionally used in the following ways:

a. Information Systems: systems that inform users about what information exists and where to find it. These may be internal to an agency or mounted on the Internet through an agency's Home Page. As part of these systems, hypertext permits instant (Internet) access to information sources no matter where that information may be located in the world.

b. Integrating Software: software that contains an integrated set of "tools" (map editors, statistics, graphics, expert system shells, etc.) which permit the user to access, assemble and use data, models, text, imagery, video, models, etc. for any purpose defined by the user. While some geographic information systems (GIS) have some of this capability, fully developed systems, such as Environment Canada's RAISON Software, are designed specifically for such tasks.

c. Advisers: special computer programs that are designed to provide advice to users. This can range from simple situations that are captured in the software and presented to the user, up to programs which use the full range of information technologies such as Expert Systems (knowledge base), neural networks (self-learning), fuzzy logic (uncertainty), etc. and which are increasingly being used in complex decision-support software. While these advanced technologies are invisible ("transparent") to the user, they often permit analysis of uncertainty in the decision process which can be very beneficial to the user. The technology of "advisors" uses much of the same techniques as information systems insofar as the advisor may lead the user to information sources that may be held anywhere on the Internet. Other "advisors" are quite self-contained and rely only on input of appropriate data by the user. An example of each is provided below.

One particular role of advisors is in the field of screening tools. As in the example below (EXPRES), a screening tool provides a consistent approach to making first-approximation judgements, usually with limited information. Screening tools, whether they be based on information systems or on biological and chemical measurements as noted earlier in this Chapter, assist the user to decide if greater attention to the issue is merited. For example, a screening tool could be used which permits a first rapid assessment of the potential for developing an irrigation scheme. The screening tool provides a first estimate of the potential (including potential impacts) and identifies those aspects which remain unclear and which require further investigation in order to provide an improved decision. The screening tool is especially useful for the non-expert and, in many instances, can save time and money by eliminating the need to send irrigation professionals with different types of expertise to every possible site. Alternatively, the screening tool can assist a single expert by providing the knowledge base of other experts as part of the database contained within the computer program.

Advisers can also lead to significant savings in labour in situations where agencies must routinely respond to technical issues. By capturing the experts' knowledge about these issues on a computer, routine responses can often be handled by a secretary thereby freeing up the time of expensive professionals.

Examples of Advisers in Water Quality: The following two examples demonstrate two types of advisers that reflect problems of water quality in agriculture. The first, a manure management system known as the "Manure Wizard", is an information system which not only assists in making a decision, it also permits the user to explore the sources of information that pertain to the recommended decision. The second, EXPRES, is a self-contained program which permits the user to explore the potential for contamination of shallow groundwater by pesticides through the use of models and pesticide databases that are built into the software.

(a) Manure Wizard

Manure management at the farm level is mainly driven by problems of water quality impacts. Manure management is often complex, involving decisions about manure chemistry, animal types, quantitative prediction, economics of manure handling, different options for disposal, spreading on land under different conditions of soil, slope and crop types, etc.. Because manure management has significant pollution potential as well as significant costs at the farm level for containment and disposal, the Manure Wizard provides the farmer with as much information as is needed to make an informed decision about his options. This Advisor, developed at the University of Guelph (Canada) for Agriculture Canada allows the farmer to interrogate the information system about any aspect of manure management under different types of agricultural conditions that are found in Ontario. The system contains relevant text as well as the ability to connect the user, via the Internet, to other sources of documentary information. The Adviser contains a "knowledge base" which helps the farmer arrive at an appropriate and cost-effective solution for manure management. The knowledge base consists of informed judgement from professionals in this field, which is captured as part of the database, and then applied to the particular problem of the user. The Advisor, once developed, provides comprehensive and systematic information for decision-making to the farmer, and requires no computer skills on the part of the user.

Figure 14 illustrates the first two "screens" (i.e. computer images) that appear in the Manure Wizard. Each screen leads the user through a series of questions and provides guidance on those issues relevant to manure management. Using "hypertext" linkages, the user can interrogate specific words, titles, phrases or issues that appear on the screen. Hypertext then immediately transfers the User to the relevant section of the Advisor or automatically connects the user to an external information source.

(b) EXPRES: The EXpert system for Pesticide Regulatory Evaluations and Simulations), was developed by the National Water Research Institute of Environment Canada (Crowe and Mutch, 1994) as a tool for quickly evaluating the potential for contamination of shallow groundwater by agricultural pesticides. Typically, this type of issue is answered by a detailed knowledge of the soil (often through drilling of cores), measurement of slope, pesticide chemistry, etc.. This type of investigation is expensive and there was a need to develop a screening technique which would allow non-experts to estimate the potential for groundwater contamination without having to go to the expense and trouble of drilling wells, making field measurements, hiring consultants, etc.

The EXPRES expert system consists of a "knowledge base" (informed judgement by experts) in this field, and which is captured as part of the database, a database of pesticide and other relevant information, and three pesticide assessment models. Using the user's available data and study objectives, EXPRES selects the most appropriate model, assists the user in construction of an input data set, initiates the assessment, and aids in the interpretation of the results. EXPRES can review pesticide and site properties, assess the potential for leaching to groundwater relative to other possible pesticides, make quantitative predictions on the distribution and migration rates of the pesticide, and evaluate the processes and factors that control the fate of pesticides in the subsurface.

Example of the first two "screens" of the Manure Wizard. These guide the user through a series of questions designed to assist the user in making a decision for the most cost-effective and beneficial way to manage animal wastes (Source: University of Guelph, Canada)



FIGURE 15 Different geographical scales that can be addressed with the EXPRES regional assessment adviser (From Crowe and Booty, 1995; printed by permission of Kluwer Academic Publishers)

EXPRES has been expanded into a regional assessment tool (Figure 15) by Crowe and Booty (1995) with three different scales of application - soil profile scale, local scale, and regional scale. The most detailed analysis is at the soil profile scale, whereas the larger scales are used as screening tools by regulators to determine the relative potentials for groundwater contamination and the need for groundwater monitoring.

Advisers tend to be designed for specific sets of conditions. Both EXPRES and the Manure Wizard are designed for application under humid temperate conditions and agricultural systems found in Canada; nevertheless, they can be adapted to include other types of climatic and agricultural conditions.


The management of water quality in agriculture is increasingly a complex and multi-sectoral problem that requires the ability to:

· predict environmental consequences;
· analyse remedial options both at the farm level and at the basin level;
· carry out cost-benefit analysis of other sectoral needs and impacts on water quality;
· identify policy options at the basin, region or national levels;
· carry out post-audit functions to determine effectiveness of the decisions, once implemented.

It is recommended, therefore, that FAO and national agricultural agencies take full advantage of the new capabilities offered by information technology that permit more consistent and reliable analysis and decision-making for complex water quality issues.

Information and decision-support systems should rely on commercial and public domain products for routine tasks. This includes database software (e.g. dBASE, etc.), existing GIS files (raster and vector) from commercial GIS systems, statistical packages etc. Where appropriate, decision-support systems should easily integrate existing software (such as SIMIS [Scheme Irrigation Management Information System], CROPWAT [Crop Water Requirements]) into the software. Such systems should be easily expandable (adding new tasks to the software) by using existing software components.

The onus is on system developers to create a computer decision-support software which is quite transparent to the User. This means that the User should be able to operate the program with a minimum of training. The software should have appropriate "HELP" tools built into the software. The User should not have to be a computer expert.

Use of water quality objectives

Water quality objectives and guidelines are widely used to determine the suitability of water quality for designated uses, including agricultural use. The impacts of agriculture on water quality, therefore, are a determinant in whether or not water is suitable for a downstream use, including downstream withdrawals for irrigation. Water quality objectives are widely used for regional water quality assessment and planning purposes, and for environmental reporting.

There is a tendency amongst developing countries to adopt water quality objectives and guidelines developed by western water quality agencies. The Canadian Water Quality Guidelines (CCREM, n.d.), for example, are widely cited. Calamari and Naeve (1994) note that the use of water quality criteria that are developed in temperate ecosystems should be used with care in African situations due to the large differences in chemical behaviour (toxicity, persistence and accumulation rates) in these different climate conditions. The same point can be made for Asian and many Latin American countries. Water quality objectives/guidelines/criteria also reflect, explicitly or implicitly, societal values and willingness to accept risk, especially for water quality objectives linked to public health. All these considerations suggest that water quality objectives developed in advanced countries may not be appropriate for developing countries. This issue is much larger than that of agricultural impacts on water quality, however the level of impact of agriculture in many developing countries is sufficiently large that agricultural agencies should have some involvement in the development of water quality objectives that are appropriate for those countries and which are realistic in terms of the ability to assess agricultural impacts on water quality for the purpose of meeting water quality objectives.

TABLE 18: Candidate pesticides for the proposed international POPs protocol












Lead compounds


Cadmium compounds




Chlordecone (Kepone)



Fluoroacetic acid & derivatives






Lindane (Hexachlorocyclohexane)








Polychlorinated terpenes

Poly-chlorinated biphenyls (PCBs)


Selenium compounds


1 Chemicals not currently used in Western Europe but identified as requiring control if they were in use. Highlighted compounds are those commonly referred to as the "Dirty Dozen".

FAO and the POPs agenda

Global contamination by Persistent Organic Pollutants (POPs) has achieved international prominence (Table 18). The importance of these chemicals in the context of non-point sources of pollution was brought out in the following fora: 1995 UNEP Governing Council; ongoing ECE negotiations on long-range transport of atmospheric pollutants (LRTAP); the 1995 adoption of a Global Plan of Action for the Prevention of Pollution of the Marine Environment by Land Based Activities; and a Nordic initiative to achieve a POPs protocol.

The implications for agriculture are that a substantial number of the candidate chemicals on the POPs list are agricultural pesticides (Table 18), some of which are still widely used in developing countries. The chemicals listed in Table 18 are taken from European, Canadian and ECE lists of pesticides that are banned or that have been identified as requiring major reductions in use due to documented environmental and/or public health consequences.


Para. 17: Acting to develop, in accordance with the provisions of the Global Programme of Action, a global, legally binding instrument for the reduction and/or elimination of emissions, discharges and, where appropriate, the elimination of the manufacture and use of the persistent organic pollutants identified in decision 1 8/32 of the Governing Council of the United Nations Environment Programme. The nature of the obligations undertaken must be developed recognizing the special circumstances of countries in need of assistance. Particular attention should be devoted to the potential need for the continued use of certain persistent organic pollutants to safeguard human health, sustain food production, and to alleviate poverty in the absence of alternatives and the difficulty of acquiring substitutes and transferring of technology for the development and/or production of those substitutes.

The agricultural sector will face significant challenges in accommodating worldwide bans or serious restrictions on certain pesticide formulations. These challenges include cost-effective alternatives, national regulations, and enforcement of importation, manufacturing and use of banned agro-chemicals. Because, at the time of writing, it is difficult to foresee what the final outcome of the various POPs initiatives will be, this recommendation deals only with the role that FAO may have to play in the negotiations that may lead to an international POPs protocol. FAO is the principal source of impartial information, both on the science and chemistry of pesticides used in agriculture that will be needed by many developing countries to effectively participate in the POPs negotiations, and on provision of informed advice to developed countries of the economics and efficacy of pesticide use in the developing world. FAO is in the position of playing an important and highly visible role as mediator and broker to the POPs process.

Pesticides in developing countries

Quite separate from the POPs agenda is the problem of pesticide use in developing countries and countries with economies in transition. The history of pesticide abuse is legend. The environmental, water quality, and public health consequences are well known. While the "Prior Informed Consent" programme of FAO and IRPTC (International Register of Potentially Toxic Chemicals) is an important step, the abuse and misuse of agricultural chemicals remains a major problem in many countries, especially in Latin America, Asia and eastern Europe. There are no easy answers or recommendations, however the issue is so important both for public health and the environment, with large off-site economic costs, that FAO needs to develop a specific action plan in the field of pesticide use. The action plan needs to include assessment, education, demonstration, chemical replacement, storage and destruction.

Actions by national governments, such as reduction or elimination of price subsidies, can have significant beneficial effects through reduced pesticide use. Combined with training in integrated pest management, reduced pesticide use can achieve both ecological (including water quality) and economic advantages at the local level.

Previous Page Top of Page Next Page