Download
1 / 12
120 likes | 263 Views
Evaluating the Extent and Relative Risk of Aquatic Stressors in Wadeable Streams throughout the U.S.A. John Van Sickle, John L. Stoddard, Steven G. Paulsen U.S. Environmental Protection Agency National Health and Environmental Effects Research Laboratory Western Ecology Division Corvallis, OR.
E N D
Evaluating the Extent and Relative Risk of Aquatic Stressors in Wadeable Streams throughout the U.S.A.John Van Sickle, John L. Stoddard, Steven G. PaulsenU.S. Environmental Protection AgencyNational Health and Environmental Effects Research Laboratory Western Ecology Division Corvallis, OR
Problem:Which aquatic stressors are important in a region’s streams? (“Stressors”-- Nutrients,excessfinesediment, degraded instream habitat, lack of ripariancover, toxicity, flow alteration, acidification, etc.) Partial Answer:-- Compare the regional extents of stressors. -- U.S. EPA’s Environmental Monitoring and Assessment Program (EMAP) designs sample surveys to estimate stressor extent. -- Stressor extent can be measured by the percent of total regional stream length in “Poor” condition for that stressor.
Example: Wadeable Stream Assessment (WSA) sampling locations, 2000-2004
Stressor extent forWSA Western region:Percent of total stream length in “Poor” conditionfor each of 8 stressors
Defining “Poor” (Most disturbed) and “Good” (Least disturbed) condition classes from stressor or response index scores at reference sites Reference-sitedistribution
However, stressor importance does not depend only on “extent”. -- We worry less about a widespread stressor, if its occurrence has little effect on biota. -- Need to also assess the size of a stressor’s effect on biota. To assess stressor effects across a region: -- Will use Index of Biotic Integrity (IBI) for stream macroinvertebrates as an example measure of stream biota condition (Good or Poor). -- If a stressor has a strong effect on stream macroinvertebrates, then “Poor” IBI conditions will tend to occur wherever stressor condition is also “Poor”. -- Use relativerisk to measure co-occurrence of Poor IBI and Poor stressor conditions.
Example: Relative risk of Poor macroinvertebrate IBI when the excess-sediment (ExSed) stressor is also Poor. Percent of stream length (Risk of Poor IBI, given Poor ExSed) = 19/23=0.83 Also -- (Risk of Poor IBI, given Good ExSed) = 18/77=0.23 Result: The risk of Poor IBI when ExSed is Poor is higher than the risk when ExSed is Good. ( Note – “Risk” is just another word for “Probability” )
Relative Risk (RR) is the ratio of these 2 risks So: “The risk of Poor IBI is 3.6 times greater in streams with Poor ExSed than in streams with Good ExSed.” Notes – -- If stressor has no effect then RR = 1. -- Use a confidence interval to express uncertainty in RR. (EMAPsoftware for the R language calculates RR and its confidence interval. Free, at www.epa.gov/nheerl/arm/ . Also, see Van Sickle et al., Environmental Management, in press).
WSA Western region: Extent of Poor condition for 8 stressors, and relative risk of stressors for Poor macroinvertebrate IBI Extent Relative Risk % of Stream Length Relative Risk ratio
Summary Challenges in using relative risk – --- May not capture joint effects of correlated stressors. --- Employs condition classes (‘Poor’ vs. ‘Good’). Advantages of using relative risk – -- Familiar language for general public. -- Employs condition classes (‘Poor’ vs. ‘Good’). -- Combines with stressor extent to assess the relative importances of different stressors.
