BY MARK BRAYKOVICH
The Cincinnati Enquirer
The nation's flood warning system is being weakened, needlessly
placing lives and property at risk, experts at two federal agencies warn.
How river gauges work
River gauges measure water levels and, in some cases, water discharges.
Some gauges are wells - 3 to 4 feet in diameter - that are housed in small concrete buildings. As water rises inside the well, measurements are taken.
Some gauges are pressure transducers, which are tubes that pump nitrogen bubbles into a stream. There is a direct relationship between the amount of pressure needed to push the bubble into the stream and the water level.
Data collected by gauges is either telemetered (transmitted by satellite), sent by phone lines, or read manually by hydrologists. The U.S. Geological Survey shares the data with such agencies as the National Weather Service, which uses it to forecast floods.
Those warnings - from hydrologists at the National Weather Service and the U.S. Geological Survey - come as the federal government prepares to shut down 200 river gauges nationwide in the coming fiscal year.
The closings, forced by previous and anticipated budget cuts, would follow the discontinuation of 363 gauge stations from 1990 through 1996.
The impact of those closings was felt in early March. Flood forecasters at the National Weather Service's River Forecast Cen
ter in Wilmington complain that a discontinued river gauge near Falmouth hampered them in analyzing last month's massive flooding. Five people died in Falmouth.
''Losing gauges is something all of the 13 river (forecast) centers have been complaining about for years,'' said Tom Adams, development and operations hydrologist at the Wilmington-based River Forecast Center, which monitors data from about 270 gauge stations in the Ohio River valley.
''Every time a gauge closes it's a concern for us,'' he said. ''We are dependent on them to know what's happening ... so that our computer models have the best data and so that we can update our forecasts.''
Even if the gauge on the Licking River at McKinneysburg, Ky., had been operational, said Mr. Adams, it is unlikely that the additional data would have resulted in earlier flood warnings or the saving of lives.
''But it would have given us much greater confidence in what information we were putting out to the public,'' he said. ''It would have provided downstream confirmation of what we were seeing upstream.''
The soaring number of gauge closings has left the U.S. Geological Survey (USGS), the federal agency responsible for collecting scientific water data, with the fewest gauges since the 1970s. States such as Kentucky and Indiana have been particularly hard hit, losing 26 and 28 gauges respectively since 1990.
And even Ohio - albeit more fortunate than some of its neighbors in losing only 11 stations since 1990 - reports its lowest number of gauges since 1941.
The cutbacks and funding shortfalls also have turned some USGS scientists into fund raisers and lobbyists - some of them increasingly mindful that no one cares about gauges except when a disaster such as the Flood of '97 hits.
''We are pushing the issue of how gauges serve as flood warning devices, and how losing them is such a problem, because that's the only thing legislatures and the public can understand,'' said Ron Evaldi, a USGS hydrologist in Charleston, W.Va.
''Unfortunately, it takes a situation like the one in Kentucky for people to care.''
A flood of cuts
While the National Weather Service benefits from river and stream gauge data, it pays nothing toward the cost of installing and maintaining gauge stations.
The agency simply piggybacks on existing gauges funded by USGS and various state and federal agencies that share the costs with USGS.
And when USGS closes a gauge station, the weather service almost always pulls out as well, though the agency could choose to keep gauges operating.
Again, money is the problem.
''We don't have the budget for that,'' said the weather service's Mr. Adams. ''The only way to have the money to do that is for the public to decide it's important enough.''
Operating the nation's roughly 7,300 gauge stations costs about $82 million annually. In fiscal 1996, USGS put up about $27 million of that, and other federal, state and local agencies the other $55 million.
Prior budget cuts forced the closing of 363 stations from 1990 through 1996, and now USGS faces more cuts. The fiscal 1998 budget calls for a $1.21 million reduction for gauges. Since those funds are part of a cooperative program in which USGS matches contributions by state and federal agencies, the result is the loss of another $1.21 million in matching funds.
The reduced funding will force USGS to close 120 stations, but the shutdowns won't stop there. That's because Congress slashed another $1 million in gauge funds in the fiscal 1997 budget.
While USGS deferred many gauge closings in hopes the funding would be restored in the coming year's budget, those hopes were dashed when the Clinton administration submitted its proposal. The result: an additional 80 to 100 gauge closings in fiscal 1998, bringing the total to at least 200.
''Clearly, without this data, the nation would experience increased losses from floods in both life and property,'' said Edward Groff, president of the American Society of Civil Engineers, during testimony in March before a House subcommittee studying appropriations to USGS.
The funding situation has become so bad that, in recent years, some states have stepped forward with matching funds only to discover USGS had no funds to match.
In fiscal 1993, for instance, $20 million in state money earmarked for gauges went unmatched by the federal government. By 1996, the gap had grown to $29 million.
There are not always willing partners in many states, however. Mr. Evaldi of USGS' West Virginia office said he spends most years scrounging for co-op partners who typically wait until the last moment to commit funds.
His total gauge program costs about $800,000 a year to run, and he needs about $200,000 to fund gauges he worries will have to be discontinued.
''It's so silly, the amount of money we are looking for, when you see how much the federal government spends,'' said Mr. Evaldi. ''I mean we're talking about half of our program going away if we can't find funding.''
It costs about $20,000 to $25,000 to install a gauge, and about $10,000 annually to maintain it. When USGS discontinues one, it tends to leave the installation intact - in the event funding becomes available in future years.
That rarely happens.
In 1994, the Kentucky USGS office in Louisville was forced to close 15 gauge stations, including the one in McKinneysburg.
''We would love to have stations throughout the whole state, but that is not economically feasible,'' said Michael Unthank, a USGS hydrologist in Louisville. ''It's all about where our priorities are.''
Among the Kentucky office's latest losses, according to a partial list of fiscal 1997 closings, was a gauge on the Ohio River near Evansville, Ind. During its 54 years, it collected loads of data for USGS and the National Weather Service.
Similarly, the McKinneysburg station had been in continuous operation for 56 years.
Losing historical data
Those discontinued stations point out another troubling trend behind the budget cuts: years of lost data.
Hydrologists note that when a gauge is shut down and its readings no longer monitored and stored, the many years of data it previously collected are rendered meaningless. In general, the older the gauge, the better its value to a hydrologist, because the long track record makes future predictions and computer models more reliable.
A 1995 study by USGS hydrologists showed that the number of gauge stations at least 50 years old had declined significantly since 1970, while the number of stations less than 20 years old had gained steadily.
Older stations also are valuable for aiding in trend analysis on such hot environmental issues as the greenhouse effect. Scientists are reliant upon lengthy records to determine whether stream flows are being affected by global warming.
''Every time a gauge is lost, we're losing a great deal of information,'' said Mr. Adams.
Computer models can still try to predict water levels at sites with no gauges, but that information is not as accurate as real-time readings.
''When you have no data, you have to take those models with a grain of salt,'' said Steven Hindall, USGS district chief in Ohio.
''And if it starts raining, those models don't help you much,'' he said. ''It doesn't help us warn residents that (a flood) is coming. And it certainly hinders our ability to provide data to anyone else.''