trends or help guide costly management decisions. One role the 
numerical model can play is to reduce the complexity and uncer¬ 
tainty resulting from the lack of long-term data sets. If we 
can develop an accurate model of the Bay's behavior during inter¬ 
vals when we have an adequate set of observations, then we can 
use this model to predict the Bay's response to the driving 
forces for intervals which are not well covered by observations. 
Long-term records and modeling are the only avenues toward 
understanding of the interannual variability of a natural 
system. This interannual variability must be addressed in order 
to be able to normalize the records from the estuary to detect 
changes that are the result of man's influences. In the case of 
dissolved oxygen, for instance, we must be able to separate the 
fluctuations due to variations in river flow (and hence, strati¬ 
fication) and increases in nutrient loadings on the Chesapeake 
Bay. 
I would like to endorse what has been a traditional role 
played by government agencies in the realm of long-term measure¬ 
ments. The National Oceanic and Atmospheric Administration and 
the U.S. Geological Survey have historically provided long-term 
observations of such variables as river flow, sea level, and 
meteorological forcing, upon which we depend greatly for insight 
into the processes of the Chesapeake Bay. Given the precedent 
and the tradition, I would like to encourage these agencies to 
initiate new long-term observations in estuaries. Moreover, the 
recent move to provide real-time or near real-time sea level and 
flow information should be commended. The Federal agencies have 
the experience and the means to provide the oversight and conti¬ 
nuity that long-term data sets require. 
Thank you. 
40 
