Data and Modeling: The Future of Water Resources Planning and Management

Abstract

Most practitioners in the water resources arena have probably already faced various problems resulting from reductions in the number of actively gauged streams in the United States. It is the result of decreased funding to support the U.S. Geological Survey stream gauging program. These reductions hinder our current efforts in hydrologic analysis and modeling, and limit what can be done in the future, particularly with the need for gauged locations with 30 or more years of uninterrupted record. Unfortunately, funding for observational systems is in peril across a broad range of fields. Equally disturbing is the movement in most countries toward the use of user fees for data, and imposing significant restrictions on the reuse of these data. One of the clearest examples illustrating this latter point comes from numerical weather prediction ~NWP! model output from agencies outside of the U.S. Here, countries are attempting to recoup the high cost of data collection or data generation from NWP modeling. While most U.S. agencies have not yet reached the level of data user fee collection found in other countries, there are significant signs that this is beginning to happen, especially as budgets shrink. As data becomes more scarce and expensive, the temptation may be to make do without the level of data that is truly necessary to correctly analyze the problems facing water resources planners and managers. As a cost saving measure a few years ago, the U.S. National Weather Service ~NWS! considered reducing the number of sites and number of times each day that its Weather Forecast Offices would launch radiosondes for atmospheric soundings within the upper air observation program. These worldwide soundings are critical for atmospheric NWP model initializations and, as Chaos research has clearly demonstrated, slight perturbations of dynamical systems can and do produce dramatically different outcomes in unpredictable ways as time progresses. With the growing concern in global climate change and climate warming, the scientific community has recognized that climate change will express itself most significantly in water resources due to the importance of water availability and scarcity, not only in economic terms, but for the survival of significant human populations across the globe. The need for believable predictions of seasonal climate variations and climate change depends not only on improvements, in the underlying scientific foundation of NWP and climate models, but also the observational basis from which these models are initialized and by which the predictions must be verified. There can be no substitute. There has been a growing trend toward increasing sophistication in deterministic hydrologic modeling over the past decade to better capture the underlying physical processes in natural hydrologic systems. State-of-the-art watershed modeling now addresses not only mass balance, but energy balance as well. This is necessary because water resources planners and managers are asking