Interannual variability at three inland water sites: Implications for sentinel ecosystems

Abstract

In the sentinel-site approach to monitoring ecological resources, relatively small numbers of locations are chosen for intensive study, each location being selected to represent a certain, preferably large, class of ecosystems. This paper examines long-term studies at three inland aquatic ecosystems in California that, although never intended as sentinel sites, nevertheless illustrate the challenges posed by sentinel-site networks. The main mechanisms of interannual variability in primary productivity or producers are described for Castle Lake, Lake Tahoe, and northern San Francisco Bay. The external forces behind the variability include climate (in the form of both snowfall and total precipitation), fishing intensity, anthropogenic nitrogen emissions, spring weather systems, intrinsic population cycles, forest fires, water diversions, and invasions of exotics. Each mechanism is associated with certain critical ecosystem features (such as hydraulic residence time) that condition a responsiveness to one or a combination of these external forces (such as precipitation); these critical features are identified for each mechanism. Three demanding conditions must be fulfilled in order for a sentinel site to function with regard to a given stressor: (1) some subset of the network must encounter the stressor; (2) at least some sites in the subset must have the critical features that cause responsiveness to that stressor; and (3) the background variability at those sites must not disguise the response to the stressor of interest. In practice, reliable extrapolation from sentinel-site networks to regional trends appears to be beyond our ecological understanding at the present time.