INTRODUCTION

The continent of Asia covers an area of 43.5 million km2, one-third of the Earth's land surface, and is home to more than half of the world's population. The vast extent of the continent – more than 10 000 km west to east – with its high mountain ranges and extensive lowlands, together with irregularities in atmospheric circulation patterns, cause uneven distributions of heat and moisture. There are permafrost zones, large river systems with areas of water surplus, and hot dry deserts.

Over most of Asia, particularly in the west and south, the distribution of precipitation is uneven, and in some areas the annual water deficit attains 1500–2000 mm. Asia is the world's major user of freshwater, and water resources are generally well developed; more than two-thirds of the world's irrigated land is located here. Total surface water resources amount to 14 400 km3/year, and water availability per capita is 13.1 m3/day, or half of the world average. Shiklomanov and Markova (1987) estimated that in 1990 water consumption for economic uses in Asia would be 2400 km3, or 59% of the world total. This represents 17% of annual streamflow, an amount that could increase to 20–25% in the near future. The development of agriculture, the main source of economic prosperity, will be limited by the availability of water. Reliable assessments of possible changes in climate due to global warming, and the effects of these changes on hydrological regimes and water resources, are therefore vital for economic planning in many Asian countries.

The influence of climate on water resources is also important for Australia, the smallest continent by area and by population.

INTRODUCTION

Historical evidence of the effects of climate variability on the hydrology of North America is found in paleorecords interpreted from a variety of sources, including ice cores, lake sediment cores, tree rings, and pack rat middens. These records help identify the occurrence, duration, and effects of climate extremes that have ranged from cold and wet periods of major glaciations, to warmer and drier periods that were ice-free. Recorded climatological and hydrological data from the past few hundred years provide more detailed measures of the effects of climate variability, including extreme events such as the drought of the 1930s and significant floods such as the recent Upper Mississippi River floods of 1993.

The effects of climate variability and change on the hydrology and water resources of different regions of North America have been investigated by a variety of researchers for several decades. The most recent concerns regarding the effects of increasing greenhouse gases on climate variability and change have expanded these research efforts (1) to better define ‘natural’ climate variability; (2) to look for the effects of greenhouse gas increases on the climate and hydrological systems; and (3) to develop methodologies to estimate the effects of future climate variability and change on hydrological systems.

A variety of empirical, conceptual and physically based models have been employed using historical data and the estimates of future climate scenarios. Early investigations of climate change used hypothetical climate scenarios based on historical records and investigators' ‘best estimates’ of possible changes in temperature and precipitation.

PROBLEM IDENTIFICATION

The subject of UNESCO's International Hydrological Programme (IHP-IV) project H-2.1, ‘Study of the relationship between climate change (and climate variability) and hydrological regimes affecting water balance components’, is relatively new, but it is of vital importance for society. Climate change and variability will affect the hydrological cycle, which will in turn affect both the distribution and availability of water resources for domestic use, for food production, and industrial activities, as well as for the production of hydropower. Other hydrology-related aspects include flood control, water quality, erosion, sediment transport and deposition, and ecosystem conservation. Most uses of water are economically important, and thus are related to socioeconomic development, as well as to public health and wellbeing and the environment, which themselves are also interrelated. Thus the concerns over the potential impacts of climate change range from the causes to the ultimate and diverse social consequences, which will vary depending on the location and the human responses. This volume addresses mainly the impacts of climate change on hydrological systems, but occasionally some other aspects are also considered.

The social relevance of this study is well illustrated in the tentative statement formulated at a meeting of the Working Group, as follows:

INTRODUCTION

Africa, with an area of 30.5 million km2, is the Earth's second largest continent, and presents the most extreme climatic contrasts. The Sahara is the largest and most extreme desert; Mount Cameroon is one of the rainiest places in the world; and the Congo River, which accounts for 30% of the runoff of the entire continent, is the world's second-ranking river in terms of discharge. This great diversity of climates and hydrological structures has been accentuated by very sharp variations over time and space.

The water resources of Africa are very unevenly distributed. The arid and semi-arid zones, which account for 60% of the continent's land area, produce only 5% of the total runoff (Margat, 1991). Although this very unbalanced situation is attenuated by such major rivers as the Niger, Nile and Zambezi, it means that some countries are dependent on their neighbours for their water – a situation that may in future lead to disputes over the allocation of these resources.

In recent geological time there have been marked climatic changes (as witnessed by fossil drainage networks and rock paintings), with sudden fluctuations in rainfall levels. In the Sahel countries the drought that has persisted since the late 1960s has greatly reduced the discharge rates of the main tropical rivers and has led to the drying up of Lake Chad almost completely (Sircoulon, 1987). In contrast, however, exceptionally heavy rainfall can also occur – as it did in 1961–64 in the Upper Nile and Congo basins – leading to an abrupt rise in the water levels of the East African lakes (Piper et al., 1986) and increasing the inflows from the White Nile to the Main Nile.

Introduction

In a seminal paper, Lancaster (1966) suggested a new approach to consumer theory. His suggestion was based on the idea that the satisfaction derived from the consumption of a good is not due to the good per se, but rather from the characteristics of the good. The hedonic analysis was developed further by, for example, Rosen (1974) and has been employed extensively for various purposes. For example, a hedonic method for estimating people's willingness to pay for a change in the provision of environmental quality characteristics has been used (with varying degrees of success) in environmental economics since the end of the 1960s. The importance of product characteristics for consumers' enjoyment was, however, acknowledged within economic research at an early stage. In an example concerned with automobiles, Court (1939) constructed price indexes that accounted for quality change by following a procedure which he called the hedonic pricing method. A crucial part of the method was the identification of measureable characteristics that relate to automobile quality.

It is interesting to note that a focus on characteristics may offer a procedure to predict the use of new products. Roughly speaking, it may be possible to describe a product which is not yet introduced into the market as a new combination of characteristics that describe already existing products. This possibility is not fruitful for some groups of products. For example, one important reason for buying a mug is probably its look, maybe the presence of a text or a picture on the mug. One or several ‘look characteristics’ that are likely to be related to people's purchase decisions are in such a case not easy to define or to measure.

Introduction

This chapter is a follow up to a previous paper in which a first attempt was made to address some of the institutional aspects of the European Commission Directive on drinking water. Whereas the previous paper, after a short theoretical introduction, focused mainly on the legal history of the Directive and its implementation in Italy, this chapter is devoted to the more theoretical question of what legal instruments can be used to avoid an accumulation of pesticides in drinking water. The current situation with drinking water is that the EU is using Directives to set strict quality standards to protect drinking-water supplies. Once these EU standards are set, the Member States have to implement them in their national systems. It has been shown that the current regulatory approach for pesticides is not functioning satisfactorily. The research by the ecotoxicology group has demonstrated that the current Italian approach of a ban on one pesticide (i.e. atrazine) is inefficient since it does not prevent an accumulation in drinking water of other pesticides. Moreover, the research has shown that the real problem with pesticides is not their toxicity, but more importantly the accumulation of potentially toxic elements in drinking water. The current regulatory approach gives the wrong incentives to pesticide manufacturers and to the users of the pesticides, the farmers. It does not lead to a reduction of toxic pesticide accumulation in drinking water.

Introduction

This chapter examines the reasons why private decision-making on the part of agricultural chemical manufacturers might lead to products with socially undesirable characteristics being sold in socially inefficient quantities. The source of the problem is market failure: some significant economic factor goes ‘unpriced’, so that economic activity is undertaken without consideration of its full impact. Two market failures will be considered. The first is the presence of externalities (both the standard story of a damage externality, in which agents considered only the private, and not social, costs of their actions; and a ‘surplus externality’, arising because profit-maximising firms consider only the marginal consumer, instead of all consumers, in their production decisions). The second is imperfect competition – the effect of the structure of the agriculture chemical manufacturing industry (a small number of large multinationals competing in the same product market) on producer's choices of chemical characteristics.

Externalities

An externality arises when the decisions of some economic agents (individuals, firms, governments) – whether in production, in consumption, or in exchange – affect other economic agents, and are not included in the priced system of commodities, i.e. they are not compensated. An alternative way of expressing this problem is to say that property rights are not assigned appropriately, so that the incidence of effect does not coincide with the distribution of legally recognised controls. A third equivalent statement is that economic agents consider only their private marginal costs when making decisions; they do not consider the total, or social (marginal), costs of their actions. It is the gap between private and social marginal costs that gives rise to the externality.

Introduction

The potential harmfulness of a chemical in the environment depends on several major properties: toxicity to organisms, bioaccumulation in tissues, persistence, mobility and distribution patterns in environmental compartments. Historically, the focus of ecotoxicology was on the first property, toxicity, and techniques were devised to enable maximum concentrations of no harmful effect to be established, usually using responses measured in whole organisms. These concentrations then formed the basis of environmental quality standards.

For many years, these studies played a vital role in the control of the major causes of chemical pollution from industry. However, in the past three decades, there has been an increasing awareness of the special importance of persistent chemicals, which can be transported and detected far from their original source; this problem was highlighted by the discovery of widespread environmental contamination by DDT and later by polychlorinated biphenyls (PCBs). Organisms over a wide area could be exposed to low concentrations of such substances for a long period of time. Ecotoxicologists then began to search for very sensitive biological responses in order to detect the effect, if any, of these low concentrations.

Considerable attention was focused on organismal effects at the cellular or subcellular level, where they first become apparent. These were often found at exposure concentrations lower than those predicted to be safe from tests on whole organisms. The ensuing scientific and political debate served only to cloud the validity of existing environmental quality standards, and indeed shed doubt on the value of ecotoxicological data in pollution prevention and control.

At the same time, the definition of pollution was changing.

Moving towards a preventive approach

Toxicology and ecotoxicology are disciplines that have developed in response to a need for information about the possible damages that might result from chemical usage. During the seventies a shift occurred from a posteriori control of chemical impacts to the prevention of this type of damage. The change in emphasis occurred first in the scientific community and then in the administrative and political spheres. As a result, many important regulations were approved for application across Europe. The essence of these regulations was to require preliminary information on the toxicology and ecotoxicology of chemicals in order to make available data needed for a preventive risk assessment of the characteristics of the marketed chemicals.

In particular, the Toxic Substances Control Act (US EPA, 1978) in the USA and the Sixth Amendment to the Directive on Dangerous Substances (EEC Council Directive, 1979) in Europe require the development of a basic set of information before a new chemical substance may be marketed. The required data set dealt with several characteristics of the substance (chemical structure, use patterns, physico-chemical properties, analytical methods, etc.) and includes toxicological and ecotoxicological tests at different levels of complexity in relation to the amount of the substance produced and the results at the preliminary levels (see Table 1).

The challenge to the scientific community was therefore: to what extent can the impacts of the chemicals be predicted by reference to this relatively limited set of data?

The problem under consideration

In July 1980 the European Commission issued a Directive on drinking-water quality (80/778/EEC) setting a maximum admissible concentration for 71 distinct parameters. One of the most strictly regulated substances in the directive was the set of chemical pesticides. The European Commission adopted a ‘practically zero’ level of permissible contamination for these substances. The limit for any individual pesticide product was set at the trace level of 0.1 μg/l; a ‘cocktail’ standard for the allowed aggregate level of contamination by all chemical pesticides was set at 0.5 μg/l. These were levels of chemical contamination that were only just detectable under then-existing monitoring technologies. The Commission's standard was intended as a clear and unequivocal pronouncment against the accumulation of chemicals within the drinking water of the EEC.

Despite this pronouncement against chemical accumulation, pesticides have been accumulating in groundwater over the past 15 years to such an extent that several substances have breached the allowed concentration in groundwater in many of the agricultural districts across the European Union (EU) (see, generally, Bergman and Pugh, 1994). This is important because two-thirds of the EU citizenry continue to acquire their drinking-water supplies from untreated groundwater, i.e. directly from the aquifers underlying their communities. In adopting its tough stance against chemical accumulation, it had been the object of the European Commission to stimulate a comprehensive strategy of pesticide management (based on agricultural, land use and pesticide management). However, the continued accumulation of pesticides in European groundwater supplies placed the EU in the position of choosing between two poor options: either the relaxation of its earlier drinking-water quality directive or the costly treatment of groundwater prior to delivery to consumers.

Introduction

Some persistence is a desirable characteristic of useful chemical products, since otherwise the chemical would degrade instantaneously into inert components of little usefulness. The objective of any purchaser of a chemical product is to acquire the activity of the product. The more persistent the product, the less often the purchaser must expend the labour and capital required to apply it. For these reasons chemical persistence is not an unmitigated ‘bad’; it is in fact an in-built characteristic driven by the demands of consumer groups.

However, the socially optimal degree of persistence is not necessarily ensured through market mechanisms, since persistence redounds to the benefit or detriment of many individuals other than the purchaser. Specifically, since chemicals that are persistent must be active within one of the various basic environmental media (atmospheric, hydrological or organic), this activity will also be experienced by the many others in contact with the same medium. Since this activity may be undesirable from the perspective of the many other persons subjected to it, they might prefer a lower level of persistence than would the individual who is making the purchase decision without taking their preferences into account. It is the internalisation of this externality that is the subject of this chapter. Policies must be designed in order to mitigate the problem of greater-than-optimal product durability when persistence ensures that people other than the consumer will feel the impacts of the product. The particular policies that we investigate here are those that will cause manufacturers and users to take the correct decisions in the design and application of agricultural pesticides.

The risk of herbicides in groundwater

Chemical control in agricultural practice has a recent history, but it has increased greatly in the last few decades. At the end of the last century in Europe, copper sulphate was already being used to control broadleaved weeds in grass crops, but it is only since the 1950s that selective herbicides have been introduced onto the market. The enormous success of these products was an incentive for research and development and this led to a number of new herbicides belonging to different chemical classes being marketed. This was followed by a big expansion of the use of chemical products for weed control worldwide. In developed countries, herbicides are used on 85% to 100% of all main crops. In the 1994 Annual Index of Weed Abstracts, 333 active ingredients are listed, and about 200 of these are widely distributed and used all over the world.

Although the environmental risk from pesticides was recognised relatively early on, the occurrence of pesticides in groundwater was not detected until much later and the major concern was directed towards DDT and other persistent organochlorine compounds. This was due mainly to lack of knowledge about the important features of chemicals movement through the soil, and gave rise to the general misconception that less persistent pesticides could not leach into the groundwater under normal conditions. One of the first references to the discovery of pesticides other than chlorinated hydrocarbons in groundwater was by Richard et al. (1975). In the late 1970s, however, the number of detections increased rapidly, along with concern from public authorities about chemical contamination of groundwater.