The bulk of the expense to implement a benthic monitoring program comes from the laboratory costs. The 

 benthic laboratory for this study charged $300 per sample for sorting and taxonomic identification. Three 

 replicate samples were collected at each station bringing the laboratory cost to $900 per station. This brings 

 the total cost of implementing the benthic index to $913 per station, excluding travel, lodging and personnel 

 costs. These costs were well within the budget allocated to this program. 



Phase 3: Response Variability 



Once the indicator has been determined to be pertinent to the assessment question, ecologically relevant, 

 and feasible to implement, then the next phase is to evaluate the expected variability in the response of the 

 indicator. Variability can arise from many sources {i.e., human error in field or laboratory processing of 

 samples, temporal variability, and spatial variability). EMAP-E has incorporated, as part of its design, 

 mechanisms to address these sources of variability in any indicator. It is important to evaluate the variability 

 in an indicator in the context of the specific assessment question. The evaluation of response variability is 

 very different for an indicator that is designed to measure the current condition at a particular site versus an 

 indicator like EMAP-E's benthic index. This benthic index was designed to measure the proportion of estuarine 

 area in the Louisianian Province that had degraded benthic communities. In this example, the assessment 

 question was aimed at estimating the spatial extent of degraded benthic communities across a large 

 geographical area (-25,000 km^ of estuarine area in the Louisianian Province). 



The data used in this evaluation came entirely from EMAP-E's efforts in the Louisianian Province estuaries 

 (northern Gulf of Mexico, from Anclote Key, Florida to Rio Grande, Texas) from 1991 to 1994. Figure 3-3 

 shows the distribution of sampling sites for a single year, 1991. The sample design, methods, results, and 

 statistical evaluations are documented in Summers et al. (1991), Summers et al. (1992), Summers et al. 

 (1993), Macauley etal. (1994), and Macauley etal. (1996). 



Guideline 8: Estimation of Measurement Error 



The process of collecting, transporting, and analyzing ecological data generates errors that can obscure 

 the discriminatory ability of an indicator. Variability introduced by human and instrument performance 

 must be estimated and reported for all indicator measurements. Variability among field crews should also 

 be estimated, if appropriate. If standard methods and equipment are employed, information on 

 measurement error may be available in the literature. Regardless, this information should be derived or 

 validated in dedicated testing or a pilot study. 



While the parsing of overall variance into specific components (e.g., measurement error) is essential to the 

 estimation of trends, our program was initially more concerned with the estimation of status. We did not 

 evaluate measurement error specifically; to do so would require a redesign of our program. We did, however, 

 determine the most likely sources of measurement error and sought to minimize this error through rigorous 

 training and quality control. Measurement errors can be introduced into the benthic data from three primary 

 sources: collection of the sample, handling and preservation of the sample, and activities in the laboratory. In 

 the field, variability in the sample would be associated with the volume of the grab, incorporation of water in 

 the sample, and human error associated with field sieving and preservation of the sample. This variability 



3-14 



