Bioassessment VII: Methodology
Presiding: T Laidlaw, USEPA; D Feldman, Montana Department of Environmental Quality
NB51F-01 08:30h
Surber and D-net: Comparison of Time-efficiency in Collecting and Sorting and of Differences in Biometrics of Benthic Insects
Surber sampler has been regarded as a quantitative equipment to collect benthic insect communities. However, D-net is thought as a qualitative sampler. Four Surber and Four D-net samples of benthic insects were collected from four types of substratum in point and non-point source pollution lotic systems in southeast Wyoming during summer and fall of 2001 and 2002. We recorded the time spent on collecting and sorting the samples in the field and in the lab. Bio-metrics were calculated based on the number of different organisms, which were identified at a genus level except for Chironomidae. T-test and ANOVA are used to analyze the data with SAS. We expect to find out whether or not time spent on collecting and sorting samples with Surber is different from that with D-net. Comparison of biometrics of benthic insects will be able to indicate whether or not benthic insects collected by Surber and D-net would provide the same results in different types of substratum. Therefore, whether difference between Surber and D-net exists will be clear.
NB51F-02 08:45h
Controlling Bias and Managing Effort in Benthic Macroinvertebrate Sorting and Subsampling
Laboratory benthic macroinvertebrate sample processing is an important yet time-consuming aspect of biological assessments. Using a 30-grid square Caton sorting tray, many protocols specify that grids are selected and organisms picked until the target number is reached. We evaluated level of effort expenditures for 5 different county, state, and national level projects comprising a total of 730 samples with target levels of 100, 200, 300, and 500 organisms. Up to 26% of the samples did not reach the project-specific target level, even when the entire sample was sorted. Reasons vary for the sparseness of samples, ranging from extremely heavy impairment, naturally low productivity, or potentially poor field sampling. In these cases it is often obvious, after only a portion of the sample has been sorted, that the target will not be attained. We describe the capturing of QC information or sorting and subsampling bias, quantified as the proportion of missed specimens recovered by independent re-checks, and suggest a performance characteristic of sorting efficiency >90%. We also propose a potential stop-sort criterion, where, if used in the projects we evaluated, would have saved approximately 1650 person-hours.
NB51F-03 09:00h
Comparing the Effects of Mesh Size on Benthic Macroinvertebrate Performance Characteristics in Montana streams
Montana's Department of Environmental Quality (DEQ) has collected macroinvertebrate data for twenty years. During this time, sampling methods and mesh sizes have been modified, though the effects of the modifications on the samples collected have not been studied. DEQ has used and continues to use both 500 and 1200 m mesh sizes. The purpose of this study is to evaluate the effects of the different mesh sizes on taxonomic diversity and metric values. Field crews followed DEQ's traveling kick sampling methods and collected samples at each site using both mesh sizes. Sixteen sampling locations were distributed throughout two ecoregions (the Mountains and the Mountain and Valley Foothills) with replicate samples collected at seven locations. We developed a suite of both quantitative and qualitative performance characteristics (precision, accuracy, bias) and directly compared them for each mesh size. Preliminary ordination results showed no significant differences between the community level performance measures. Preliminary metric analysis showed that the 1200 m mesh captured a greater abundance and diversity of caddisflies (Trichoptera) than the 500 m mesh. This study will determine if data collected using different mesh sizes can be aggregated for development of bioassessment tools and will help DEQ implement consistent statewide sampling protocols.
NB51F-04 09:15h
The Effect of Subsampling Effort on Macroinvertebrate Community Variables in Streams of the Interior Highlands (Arkansas, USA)
Bioassessment of stream invertebrate assemblages is increasingly used to assess the effectiveness of silvicultural best management practices. An important challenge in these studies is determining the size of a subsample that is representative of the assemblage, but remains time- and cost-effective. We examined community characteristics calculated from 100- and 300-organism subsamples of 90 benthic invertebrate samples, to determine whether the larger subsamples provided enhanced ability to discriminate between seasons and ecoregions. Samples were collected from five streams in two ecoregions during two seasons. Total richness and EPT richness were consistently greater in the larger subsamples in both winter and spring for each study stream. However, fourteen community characteristics that were based on relative abundances of various groups rarely differed significantly between 100- and 300-organism subsamples. Seasonal variation in all characteristics was equally evident in subsamples of both sizes. Similarly, differences in assemblage characteristics between ecoregions were as evident in the smaller subsamples as the larger ones. Our results did not indicate that the larger subsamples enhanced the ability to discriminate between samples taken from different seasons, or from unimpaired streams of these two ecoregions.
NB51F-05 09:30h
Wanted Dead or Alive?
A European standard has recently been developed for the sampling of macroinvertebrates in streams. The method for sorting macroinvertebrate samples, however, is still a major point of discussion. The goal of this study was (1) to determine whether preserved and unpreserved samples show significant differences in metric and bioassessment results and (2) to compare the cost efficiency of both protocols. Data from three streams in the Netherlands were collected. In each stream six replicate samples were taken with a pond net. A replicate can be described as a composite sample from several habitats at a site. Three replicates were fixed in formalin directly after sampling. The remaining replicates were not fixed. All samples were transported to the laboratory for sorting. The samples with live organisms were kept in a refrigerator until sorting was undertaken. Sorted organisms were preserved in ethanol and later identified to species level. Of the 58 metrics examined, only nine showed significant differences between protocols. Assessment results differed in only one out of six cases. No significant differences in the total time (costs) for sample processing were detected. Overall the effects of differences between replicates were larger than the effects due to differences in protocol.
NB51F-06 09:45h
Comparability of Benthic Macroinvertebrate Sampling Methods in Montana Streams
There are several benthic macroinvertebrate sampling techniques currently being used by the Montana Department of Environmental Quality (DEQ) that evaluate instream water resource quality. A traveling kick or a jab with a D-frame net collects most samples. DEQ monitors a smaller proportion of sites using Hess Samplers. Also used in the state are the sampling methods of the U.S. EPA Environmental Monitoring and Assessment Program (EMAP) and Regional Environmental Monitoring and Assessment Program (REMAP). Both EMAP and REMAP use a D-frame net, distributing sampling effort throughout the target reach along a series of transects. The purpose of this study is to evaluate the effects of these four different sampling techniques on the samples gathered, in particular, on taxonomic diversity and metric values. This evaluation will help DEQ accomplish its ultimate goal: to implement consistent statewide sampling protocols. We developed a suite of both quantitative and qualitative performance characteristics (precision, accuracy, bias) for each of the methods, and directly compared them among the protocols. The level of comparability among methods was characterized and the acceptability of each method for use by Montana DEQ was determined.