4. Synthesis of Information and Observation Requirements

Workshop participants identified five major reasons for requiring hydrological information, described briefly in sections 4.1 to 4.5 below.

4.1 Improved Climate and Weather Prediction

Main objectives of climate and weather prediction are to improve the accuracy of weather predictions with the most detailed spatial resolution feasible and to substantially improve the capability for seasonal forecasting at regional scales. Energy exchange between the ocean, land surface and the atmosphere is the major driver of the weather engine. Therefore hydrological observations on the highest possible spatial and temporal resolution are required to feed and validate next generation forecasting and prediction models. The following observations would greatly improve the reliability of NWP and climate model predictions (refer also to section 3.1):

• Near real time measurements of river runoff near to the mouth of major world rivers for coupled climate model input, and for the testing of global thermohaline circulation in these coupled models.

• Atmospheric observations such as precipitation. The technical definition of near real time hydrological observations needs to be fitted to meet the actual data requirement of specific numerical models and anticipated products, such as forecasts

• River water sampling for the off-line determination of the isotopic composition as a test of coupled models to reproduce the water cycle on land. The GNIP network is a natural starting point for these activities

• Soil moisture fields as starting fields for weather forecasting (especially medium range) and climate variability predictions, distributed via GTS to forecast centres

• Start to develop satellite remote sensing of hydrological variables, e.g. soil moisture and river runoff. Use existing satellite data (e.g. from SSMI) on snow cover and snow water equivalent.

4.2 Characterising Hydrological Variability to Detect Climate Change

The workshop participants noted the need to understand and to characterise the variation of hydrologic variables in response to climatic variations in order to be able to detect trends, that is, changes in the fundamental nature of the climatic status quo. However, such changes are not just ones of anthropogenic origin, as is clearly highlighted by the existence of past ice ages as well as more moderate climatic excursions such as the medieval warm period which could not have a human origin. The ability to discern trends in the context of expectable variations is necessary not only for reasons of scientific knowledge, but also for long term infrastructure planning. Thus, the one primary user of such information would be the UNFCC through the IPCC process and its contributing scientific collaborators, national and regional governments, and international decision and risk takers such as insurance companies.

The following observations are particularly important for climate variability studies:

• Monthly runoff series for the top few hundred catchments (in size) around the world, if possible with naturalised flows

• Daily runoff series for a few hundred smaller natural catchments (~1000 km² in size) used for research spread over the globe

• Long time series of hydrological records

• Comprehensive catalogues of metadata that explain the data sets within a global hydrological network.

An important current science issue concerns the causes of the increased frequency and intensity of ENSO events during recent decades: do they represent a long-term natural fluctuation or are they of anthropogenic origin? The answer cannot be given yet. However, it may be possible to obtain it by using measurements of the isotopic composition of the water molecules in annual natural deposits like ice-cores, lake varves, corals and tree rings. Because these data can extend the records back in time into the pre-instrumental period, if circulation anomalies like ENSO or the North Atlantic Oscillation (NAO) leave any imprint on isotope patterns. First exploratory measurements to the GNIP database and using the GNIP database point to the El Nino patterns in South American isotope data. For improved transfer functions between conventional hydrological data and isotope patterns derived from point measurements of both data types we need more monthly mean isotope data at more precipitation stations and also isotopic composition of river water, preferably at gauging stations. In addition, isotopic composition during precipitation events has to be measured to test the reliability of the mean monthly data.

4.3 Developing the Ability to Predict the Impacts of Change

The goal is to understand the process of change, both for scientific purposes and to establish the capability for implementing mitigation measures as necessary. Indeed, some argue that change can and does occur and that its consequences need not be catastrophic (e.g., von Storch and Stehr, 2000), so that an ability to assess the need for mitigation or lack thereof is critical.

A major product to be derived from a global hydrological network will be information on water availability and distribution. Climate variation has the potential to significantly alter the natural distribution of water-resources world-wide. Changes in precipitation and shifting patterns of water distribution can lead to increase of floods, water shortages and drought. In addition to water quantity, water quality aspects are an important issue , both from societal perspective and to maintain healthy ecosystems. Indeed, freshwater limnological ecosystems are known to be responsive to climate change impacts, but the magnitude of effects on these systems is just beginning to be quantified (McNight et al., 1996).

An important component of climate impact assessment concerns the transport of materials (surface and subsurface) from land to oceans. The movement of carbon and nutrients has potential impacts on coastal margins as well as in sensitive ocean regions.

The following observations are particularly important for change impact studies:

• Daily runoff series for a few hundred smaller natural catchments (~1000 km² in size) distributed over the globe, together with precipitation and other data permitting the study of hydrological processes in specific regions

• Monthly or seasonal information on biogeochemical materials transported from land to the ocean with appropriate spatial and temporal resolution to characterise global as well as regional impacts

• Multiyear ground water, surface water and water use time series to characterise water availability

• SW flux values to anticipate change in storage in natural and artificial reservoirs.

4.4 Assessing Water Sustainability as a Function of Water Use Versus Water Availability

The concept of sustainability implies that a resource is used in such a way as to remain available to that use for an unlimited time period with no detrimental environmental, economic or social consequences.

In order to assess sustainability of human water use and to model water scarcity in the context of climate change, information on water availability and water use are needed. (Döll et al., 1999; Vörösmarty et al., 2000). As for climate impact assessment, water quality is critical information because it determines is the utility of water available for the requisite use. Thus, water quantity observations that consider availability, distribution, location, and scarcity need to be co-ordinated with water quality aspects of systems that address human health, ecosystem viability, water use requirements, and the transport of materials. The following observations and data are particularly important for water availability studies:

• Area (and volume) of lakes and wetlands (time series);

• For assessing the role and impact of artificial reservoirs: location, volume-area function and management record of reservoirs, river discharge directly upstream and downstream of (large) lakes and reservoirs, and naturalised flows;

• For assessing the impact of water quality: location and other data for wastewater treatment plants, water quality parameters (in particular nutrients) in rivers, and source emissions of relevant substances, particularly nutrients;

• For assessing water use and its impact: withdrawals and consumptive water use differentiated according to source (groundwater, surface water) and user sector (agriculture, households, manufacturing, cooling, etc.) and administrative units (e.g., counties, provinces,..).

4.5 Understanding the Global Water Cycle

Isotopic tracers provide a mechanism for assessing our ability to understand the dynamics of the water cycle by allowing us to account for the flux of water between natural reservoirs (clouds, humidity, surface storage in lakes, surface channels, soil, plants, ground water, etc.), and by assessing the processes through which the water molecules proceed in the water cycle. Stable isotopes of oxygen (Oxygen-18) and hydrogen (Deuterium) have long been known to vary in precipitation and atmospheric moisture in response to meteorological conditions and moisture sources (Rozanski et al, 1993; Araguas-Araguas, 2000; Gat, 2000). Isotopic methods using tracers such as Tritium can be used to study catchment-scale dynamics, both the partitioning of water between surface and groundwater components and the residence time of water within a watershed. (Michel, 1992.) While an isotope network exists (Global Network for Isotopes in Precipitation, GNIP), few global or continental scale data sets of isotopes in runoff have been obtained so far. (An exception is the work of Coplen and Kendall, 2000.)

Concurrent observations of variability in precipitation, surface water and groundwater, as well as information on their isotopic composition will more fully test our ability to determine water balance and the water and energy exchange processes, and hence better understand both climatic processes and the hydrologic cycle.

4.6 Summary

Table 1 provides a summary overview of the importance and usefulness of individual hydrological variables in relation to the five thematic areas. The type of use of each variable is also given. The table provides a link to the discussion of individual variables that follows in section 6.

Table 1. Summary Table of Applications vs. Hydrological Variables

e= essential; d= desirable; v= validation; i= input