Complete Definition of the Emergency

In general, droughts, floods, and tropical storms produce the greatest number of water/weather-related disasters. A list of types of these disasters is given below:

• Floods (rain, snow, rain/snow).

• Land slides/mud flows/debris flows.

• Earthquakes (water-related failures- dams, canals, tunnels, soil liquefaction).

• Volcanoes (lahars).

• Saltwater intrusion.

• Dam/levee/canal failures.

• Wind-related (seiches, tornadoes, coastal flooding, hurricanes, tropical storms, dust and sand storms, ocean waves).

• Sedimentation/sediment transport (reservoirs, rivers).

• Ice jams.

• Avalanches (granular/powder).

• Climate change (greenhouse effects).

• Droughts/food shortages/famine/water shortage.

• Tsunamis (earthquake-related).

• Heat waves.

• Cold waves (snow, ice, blizzards).

• Toxic materials (water-related).

• Environmental disasters and sustainable development of small islands

Table 1 lists some major flood disasters since the year 1400.

The number of agencies and conferences, both national and international, involved in water/weather assistance, research, and development is very large. Unfortunately, cooperation between this large number of agencies is difficult to achieve and frequently is poor. A partial list of agencies is given in Table 2.

Hydrologic Forecasts

For 2000 years, levee systems and flood control reservoirs have proved to be a major deterrent to flood damage. Unfortunately, flood control reservoirs and detention basins have restricted volume, and levee systems are constructed to a finite height and produce a limited floodway capacity. In addition, operation of detention basins and flood control reservoirs and maintenance of levee systems depend on accurate prediction of floods. The primary steps in the development of an efficient and time flood prediction system include:

• Data automation.

• Development of rapid communications.

• Accurate prediction algorithms.

• Timely forecast dissemination.

• Quick response civil defense programs.

Recent enhancements in computing capabilities, modernized data collection, improved communications, better prediction algorithms, improved structural technology, flood-proofing, restrictive zoning, and better water resources management have improved the effectiveness of hydrologic forecasts. The development of greater computer capabilities has allowed the development of conceptual models that allow rapid computation of soil moisture accounting over large drainage basins. A real evaluation of soil moisture is critical to accurate estimation of runoff from excessive rainstorms. Rainfall can now be determined with considerable accuracy through use of automated and radio-transmitting rain gages, radar estimates of rainfall, and satellite-inferred rainfall amounts. Routing procedures of flood flows have improved considerably due to the use of more rigorous dynamic techniques that solve the governing St. Venant equations.

Probably one of the important improvements being incorporated into modern hydrologic forecasts is to utilize Quantitative Precipitation Forecasts (QPFs). Meteorological forecasts based on recently developed numerical models (especially the ETA model developed by the National Weather Service, USA) are as good at the end of three days as previous QPFs were for one or two days - a marked improvement. These improved forecasts help to eliminate the past tendency to "under forecast" major flood events. A more detailed discussion of weather forecasting is presented below.

Weather Forecasting

The theoretical and practical aspects of forecasting weather date from the early 1900s, some 30 years before the beginning of hydrologic forecasting. During the 1920s, the mathematical and physical concepts related to numerical weather prediction were proposed. However, it was not until after the development of electronic computers in the 1940s that it became possible to solve, in real time, the pertinent equations to forecast weather globally, both time and accurately. The tremendous increase in speed of numerical processing has been accompanied by an equally large decrease in computer costs. This has opened the field of numerical weather prediction to meteorological services lacking resources to purchase supercomputers. In addition, the development of the internet has enable services to obtain meteorological data available through the WMO WWW program.

A component of World Laboratory Project Land-1, supported by Project MEDUSE, included the preparation of quantitative precipitation forecasts (QPFs) for the Yellow River basin in east China. Testing of several numerical models indicated that the ETA model (24 L/50 km) (atmospheric levels/grid spacing) developed by the National Weather Service (NWS), USA, and used operationally at the National Center for Environmental Prediction (NCEP) in Washington, D.C., produced the best QPFs. Six-hour resolution, 72-hour precipitation forecasts were prepared for two months in 1994 and 1996 at the Centre Ettore Majorana for Scientific Culture , in Erice, Italy, and transmitted daily to the Chinese Academy of Sciences in Beijing and the Yellow River Conservation Conservancy in Zhengzhou. The results were remarkably good. Daily verification was based on 600 precipitation stations in China with a gage density of one gage/10,000 km², compared to the forecast grid of one forecast point/2500 km².

Recent(1999) forecast verifications in the United States of the ETA model have shown continued improvement using a 50 L/29 km model. Another model, the Pennsylvania State University/National Center for Atmospheric Research Meso-scale Model 5(MM5), has been tested recently by the Atmospheric Sciences Department at the University of Arizona, USA, and has produced good results. The MM5 model has been tested using 27 vertical levels and a high resolution(7-km) grid.

Another forecast model investigated extensively by World Laboratory Project Land II, on drought and desertification in China, was the RAMS model(regional atmospheric modeling system) developed by Colorado State University. That model was coupled by the University of Torina/Alessandria to a land-surface-process model(LSPM) for predicting and simulating the exchanges of heat and moisture from soil to the atmosphere for determining limited area atmospheric circulation. The RAMS model is compatible with both the ETA and MM5 models and could be integrated into forecasting schemes for the purpose of analysis of convective exchange of heat and water in arid regions. In particular, the Land II research has achieved meaningful improvement in predicting rainfall anomalies in the different climatic regions of China.