Premium
A Phosphorus‐Based Fish Kill Response Function for use with Stochastic Lake Models
Author(s) -
Mericas Constantine,
Malone Ronald F.
Publication year - 1984
Publication title -
north american journal of fisheries management
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.587
H-Index - 72
eISSN - 1548-8675
pISSN - 0275-5947
DOI - 10.1577/1548-8659(1984)4<556:apfkrf>2.0.co;2
Subject(s) - environmental science , fish kill , phosphorus , fish <actinopterygii> , function (biology) , nutrient , statistics , ecology , hydrology (agriculture) , fishery , algal bloom , meteorology , mathematics , biology , phytoplankton , geography , chemistry , geotechnical engineering , organic chemistry , evolutionary biology , engineering
Twelve summer fish kills were observed and documented over a 3‐year period in the hypereutrophic University Lakes System in Baton Rouge, Louisiana. Examination of water quality data associated with these kills revealed an apparent threshold level of 0.400 mg/liter total phosphorus below which no kills were observed. The data indicate a uniform frequency of occurrence of summerkills of 3% when total phosphorus levels exceed 0.400 mg/liter. Above this critical value, physical factors such as temperature, light, and wind apparently become the controlling parameters in algal growth. Optimum weather conditions promote blooms that subsequently collapse when the conditions change, resulting in oxygen depletion and suffocation of fish populations. This observation may be formalized in probabilistic terms as a probability density function describing the risk of a fish kill at any given nutrient level. The value of such a representation lies in its use in conjunction with a stochastic phosphorus model for estimating overall summerkill risk. The region of intersection of the model projection and the kill response function defines the risk to the system. An example of this application revealed an order of magnitude reduction in risk (from 1.9 to 0.2% per day) in comparing two potential lake management alternatives. The approach is very suitable for the incremental cost analysis of management options. Regional data are needed to verify the kill response function and develop a broadly applicable function based on geographical and morphological lake characteristics.