Background up to the twentieth century

The Aral Sea is the largest inland drainless water body with specific marine and lacustrine features located in the Central Asian deserts – in the Turanskaya Lowland, near the eastern edge of the Ustyurt Plateau. The water and salt budgets, the level and areal extent of the sea, the water salinity level, as well as other characteristics of the sea, are completely determined by the streamflow of Central Asia's two major rivers, the Amudarya and the Syrdarya. The amount of surface water flow into the Aral Sea depends on (a) the impacts of climate-related fluctuations of the natural water supply of these rivers and (b) on the steadily growing anthropogenic demands on the rivers' waters during the twentieth century (e.g., the consumptive withdrawals of river water mainly to meet the demands of irrigated farming).

Changes in the regime of the Aral Sea are determined primarily by alterations of two of its key characteristics: its height above sea level and the average level of mineralization of its water (i.e., salinity). The sea's level depends on the volume of water that accumulates in its basin. The average salinity of the sea's water is governed by the relationship between the mass of salts dissolved in the water and the volume (rather, the mass) of sea water.

Paleo studies indicate that there have been significant, age-old cyclical alterations of Aral Sea level and salinity throughout its history. For example, the range of sea level fluctuations during the Holocene apparently exceeded 20 meters.

The environmental disaster in the Aral Sea basin is a direct result of societal overuse of the water resources of the basin's two major rivers, the Amudarya and the Syrdarya. Societal activities are responsible for serious reductions in both the quantity and the quality of river flow and for negative impacts on the major components of the region's ecosystems which are directly dependent on the availability of surface water. Thus, an assessment of the region's water resources, including changes in their use, is necessary to identify the impact of anthropogenic factors on river flow and to identify hydrological measures that could improve environmental conditions in the basin.

The Aral basin lies in the heart of the Eurasian landmass, far from the direct influences of the oceans. It therefore has a continental climate (i.e., marked seasonality) and an abundance of warm and sunny days. The major part of the basin is covered by the Turan Lowland deserts bordering on the Tien Shan and Pamir-Alai mountains. Glaciers and snowfields are the main sources of water feeding the region's two major rivers – the Amudarya with a catchment area of 1.1 million km2, and the Syrdarya with a catchment area of 0.44 million km2. Unlike the mountainous part of the basin, which is the zone of formation of river flow, the lowlying flat land is a zone where water resources are dissipated as a result of evaporation from both irrigated and natural areas (Shults, 1965).

Irrigation farming ever since the distant past has been the main economic activity in the Aral Sea basin.

Introduction

More than a hundred papers have been published within the past 50 years about climate and atmospheric processes in the Aral Sea basin. A thorough review of them would require the publication of a separate book. Thus, in this chapter only one key aspect of climate in the Aral Sea basin is investigated: a diagnosis of regional climatic change and creeping environmental change.

The following sections are presented in general terms: natural changes in regional climate characteristics, temporal and spatial scales of change and their interactions, the rate of climate change and local anthropogenic impacts on climate. The data presented in this chapter are important for a retrospective identification of threshold levels in the development of gradually accumulating ecological changes in the Aral basin.

Basic works dedicated to the diagnosis of climate changes in the Aral Sea basin can be divided between those of the 1960s and 1970s and those of the 1980s. The steady decline of the Aral Sea level became obvious in the 1960s, thereby stimulating scientific studies. Another reason for the growth in the number of Aral-related publications was the fact that, by 1970, spatially detailed meteorological observation data in the Aral basin had already covered a period of over 30 years and were in need of scientific analysis. The development of the classification of types of the synoptic processes of Central Asia (Bugaev et al., 1957) and the classification of macrocirculation processes as applied to the Central Asian Kazakstan region (Baidal, 1964) were completed in the 1950s.

The first time I managed to see the Aral Sea and the Amudarya was in 1952, when large-scale studies on the northern route of the proposed but never developed Main Turkmen Canal were being developed. The Amudarya's delta with its blue- and green-colored lakes and vegetation contrasted sharply with the grayish, yellow-green desert areas of the Kynyadar' inskaya ancient alluvial deltaic plain and the saline soils (e.g.,solonchaks) of the Sarykamysh depression. Nothing noticeable was taking place in the natural environment at that time that warned us of the adverse environmental changes that would be observed after 1961.

In fact, I had noticed the first alarming symptoms of irrational use of water and of the adverse changes in the vegetational mix of desert landscapes even earlier, in 1958, while carrying out field studies of the Karabil'skaya fresh water lens in the southeastern part of the Karakum Desert. At that time the depressions between the ridges in the desert sands were beginning to flood as a result of the discharge of Amudarya water into the Karakum Canal which was then under construction. My concern about these environmental changes intensified in 1963–64, when I observed the filling of the Sarykamysh depression with waste water released through the Daryalyk channel. An eyewitness (Dr G.S. Kalenov) confirmed that, during the first year of water discharge into the Sarykamysh depression, the water disappeared into and filled with a rumble numerous underground karst cavities. Only after those underground cavities were filled did Lake Sarykamysh begin to appear in the Sarykamysh depression.

Turkmenistan occupies an area of 491200 km2. Eighty percent of the area is Karakum sandy desert, 17% is mountains, water area, and waste land. Only 3% of the area is occupied by natural and man-made oases. Turkmenistan has some fertile land and modest water resources located primarily in the eastern part of the country.

Prior to construction of the Karakum Canal, the 110–160000 ha of irrigated lands were dependent on local runoff with the water supply being unstable and insufficient. In western Turkmenistan and the Pre-Kopetdag regions, water was insufficient not only for irrigation but for domestic use. Agriculture in southern Turkmenistan was unstable as water resources were frequently on the verge of collapse. The poor water supply in those regions restricted development of other industries, primarily oil, natural gas and chemicals.

To promote Turkmenistan's economic development, it was essential to supply water to its internal regions from the Amudarya, a river with an average annual runoff of 63.8 billion m3 near the town of Kerki. The problem was successfully resolved by the construction of the Karakum Canal. The canal enabled this Republic to irrigate lands and desert pastures, supply water to cities and industrial centers, create an agricultural infrastructure near the cities, establish recreational areas, develop inland fisheries, and create a navigable waterway. Karakum Canal construction was largely responsible for promoting the development of the republic's productive forces and contributed to its economic development. Agriculture in the Murgab, Tedjen, and Pre-Kopetdag oases was dramatically improved. Prior to the beginning of the Karakum Canal operation in 1958, the total irrigated area in Turkmenistan was about 166000 ha.

This final report of the Working Group of UNESCO's IHPIV project H-2.1 provides an overview of ongoing activities in the ‘trans-science’ domain that is the concern of hydrologists, water resource engineers, climatologists and meteorologists. The members of the first two communities need inputs from the latter two, but climate change and variability is in turn determined by changes in hydrological conditions. The interaction between climate and hydrological conditions is also affected, either directly or indirectly, by other factors such as population pressure and changes in land use.

This chapter presents the final conclusions of the Working Group, and their recommendations for research and measures in the future. It is hoped that the implementation of these recommendations will help to improve the practical utility of the results of large-scale modelling at the catchment scale for water resource managers and policy makers.

CONCLUSIONS

The conclusions of the Working Group are discussed under the following headings:

1. (a) the use of paleoclimate scenarios;

2. (b) the use of GCM scenarios;

3. (c) hydrological models and climate change;

4. (d) the uncertainties associated with GCMs and hydrological models;

5. (e) field experiments to improve GCMs;

6. (f) climate change and water resources management; and

7. (g) the outcomes of the IPCC process.

The use of paleoclimate scenarios

Paleoclimate scenarios are used to provide analogues of changes in climate that have actually occurred in the past, even though the causal factors may not be known and therefore can not be easily incorporated into GCMs.

INTRODUCTION

Atmospheric and hydrological models provide a framework within which the relationships between climate and water resources can be conceptualized and investigated. These investigations can range from simple one-way couplings using atmospheric model outputs as adjustment factors for measured inputs to hydrological models, to more complex, fully coupled atmospheric and hydrological model applications that incorporate feedback mechanisms between the two systems.

The scientific literature contains a large number of reports on the variety of atmospheric and hydrological models that have been used to investigate the effects of increasing greenhouse gases on climate, and the resulting impacts of any associated changes in climate on local and regional hydrology. Current atmospheric models have generally been shown to have a limited ability to simulate present climate conditions, and thus a large degree of uncertainty exists as to their applicability for impact assessments. Similar uncertainties have been identified in the applicability of the wide range of hydrological models. The purpose of this chapter is not to provide an extensive review of the findings of these reports, but to characterize the state of atmospheric and hydrological modelling for use in simulating the effects of climate change and current climate variability. Methodologies of modelling and climate scenario development are reviewed, their deficiencies are discussed, and additional research needs are identified.

HYDROLOGICAL MODELS

‘Hydrological modelling is concerned with the accurate prediction of the partitioning of water among the various pathways of the hydrological cycle’ (Dooge, 1992).

INTRODUCTION: PURPOSE AND OBJECTIVES OF THE IPCC

In the near future significant changes in the global climate can be expected, with an increase in mean air temperature of 3–4°C. For the northern temperate regions and high latitudes in particular, these changes are likely to affect a wide variety of physiographic features over vast regions, human living conditions, socio-economic structures and development, and natural ecosystems. The effects of a rise in global air temperature could be devastating: extensive melting of the polar icecaps and the resultant rise in sea level would impact coastal areas, and changes in atmospheric circulation would affect agricultural productivity and food supplies, as well as water resources in many countries.

In the early 1970s some scientists first warned of the possible impacts of human activities on the global climate (Budyko, 1972) and predicted a warming of 1.5–2°C in the coming decades. At that time, however, few climatologists supported the view that significant climate warming would result from increased concentrations of CO2 in the atmosphere. On the contrary, it was widely believed that climate cooling would occur, and such a cooling trend was observed during the 1970s, according to data from the world meteorological network. Nevertheless, the first World Climate Conference (WMO, 1979) concluded that the human impacts on climate were serious, and that the problem required further study, but noted that science could not give a definite answer to the question of whether climate warming or cooling would occur.

The situation changed rapidly in the 1980s, when observation data showed a sudden rise in global air temperature.

INTRODUCTION

Very few studies of the impacts of climate variability and change on water resources systems have been reported in the literature. Notably, the only published report dealing specifically with South America presents the efforts of the US Environmental Protection Agency in cooperation with the Federal University of Rio Grande do Sul, Brazil, to analyze the impacts of climate change in the Uruguay River basin. Another report (Stakhiv et al., 1993) describes the results of an impact study in the Orinoco River basin in Venezuela. Unfortunately, the results of these studies were not available at the time of preparation of this report, nor those of a Russian–Argentinean effort in Argentina.

Due to its strategic importance for the world's climate, the Amazon basin has been the subject of many studies of hydroclimatology, involving Brazilian and other international institutions. These early efforts included the ARME and ABRACOS projects, developed by institutions in Brazil and the UK, whose objective was to assess the forest water consumption and exchange with the surrounding atmosphere. A major scientific experiment involving all the Amazonian countries and the international community is now being planned: the LAMBADA project (now part of the broader LBA project) will focus on the hydroclimatology of the region and will provide important data for future studies of the impacts of climate change on tropical regions.

This chapter describes the above studies and the planned experiment, and presents their main results.

INTRODUCTION

The droughts of 1988–92 and the floods of 1994 and 1995, both of which affected many parts of Europe, served as a reminder that climatic and hydrological variability still have significant economic and social impacts, despite decades of investment in a wide variety of water and river management schemes. Inevitably, the question of whether these ‘extreme’ events were signs of global warming has frequently been raised. It is, of course, too early to draw any definitive conclusions, but the recent floods and droughts have shown the sensitivity of water resources in Europe to change.

There have been several studies into the potential effects of climate change on hydrological characteristics in Europe, using unfortunately a wide range of scenarios and different methods of analysis. This chapter reviews many of these studies and summarizes some of the experiments that have been undertaken in Europe to explore the processes relating climatic variability to hydrological behaviour. First, however, it is necessary to summarize European hydrological regimes.

HYDROLOGICAL REGIMES IN EUROPE AND THEIR VARIABILITY OVER TIME

In the most general terms, hydrological regimes in Europe can be divided into two types: regimes dominated by rainfall, and regimes dominated by snowmelt. Rainfall-dominated regimes, with maxima in the winter and minima in late summer, occur in the west and south, whereas snow-dominated regimes, with maxima in spring and minima in summer or winter, are found in the north and east. In practice, of course, the picture is considerably more complicated than this (Krasovskaia et al., 1994).