Progress 07/01/02 to 06/30/07
Outputs
OUTPUTS: In this project we were successful in developing several tools whereby the amount of in-laboratory processing of sampled benthic macroinvertebrates can be greatly reduced. This was done by using a special measure of diversity - the sequential comparison index, and by coupling it with some special statistical tests - the sequential sampling procedure. The sequential comparison index, or SCI, is one that conforms to a known statistical distribution (binomial). Values for the SCI range from nearly 0.0 to 1.0. An SCI value, which is less than 0.2, and is calculated from sampled benthic macroinvertebrates, strongly indicates disturbance from an unnatural process, such as a pollutant. An SCI value that is 0.8 or greater strongly indicates an undisturbed stream or river. These values, along with knowledge of the statistical distribution of the SCI, allowed us to develop several sequential sampling tools. In 2002, we selected two binomial sequential sampling tools for further testing on 83 samples of benthic macroinvertebrates that were collected from various streams in and around southeast Wyoming (mostly). The savings in time and number or organisms examined ranged from 42.4% to 93.3%. We then investigated how selected binomial sequential sampling models could potentially influence the outcome of the sequential comparison index, or SCI, and attempted to determine how the estimate of the SCI is influenced by two levels of taxonomic expertise. For each of the selected binomial models, both non-technical as well as highly technically trained individuals sorted through samples of benthic macroinvertebrates to calculate the SCI. Non-technically trained individuals were given some basic instructions such as like - appearing organisms are to be considered as the same run, even though they are of different size while non - like - appearing organisms indicate the beginning of different runs. A repeated measures analysis of variance test indicated that, indeed, the non-technically trained individuals estimated the same SCI values as did the highly technically-trained individuals. PARTICIPANTS: Two graduate students were trained on this project. Mr. Ryan Stitt and Mr. Dayong Wu. TARGET AUDIENCES: Target audience is grass roots organizations interested in biological monitoring of streams. PROJECT MODIFICATIONS: Not relevant to this project.
Impacts
We expect this work to stimulate grass roots organizations that perform routine biomonitoring to investigate either adapting or creating their own sequential sampling programs
Publications
- Wu, D., and D. Legg. 2007. Structures of benthic insect communities in two southeastern Wyoming (USA) streams: similarities and differences among spatial units at different scales. Hydrobiologia 579: 279-289.
|
Progress 01/01/06 to 12/31/06
Outputs
OUTPUTS: ddfwsdfweffe PARTICIPANTS: test test All users who request entry to the system must prove they are who they say they are (Authentication). This applies inside CSREES as well as outsides. Who the users are determines what information they have access to (Authorization) TARGET AUDIENCES: test All users who request entry to the system must prove they are who they say they are (Authentication). This applies inside CSREES as well as outsides. Who the users are determines what information they have access to (Authorization) PROJECT MODIFICATIONS: test All users who request entry to the system must prove they are who they say they are (Authentication). This applies inside CSREES as well as outsides. Who the users are determines what information they have access to (Authorization)
Impacts
wrgrwhhwghreh, the number of characters and spaces should not exceed 3200
Publications
- Mark James 2007 did a very good job finding new products which will lead to healthy life
- James Peter 2006 was the first to find out on bio fuel
|
Progress 01/01/05 to 12/31/05
Outputs
Progress in 2005: In 2005, we began statistically analyzing data and writing a paper to be submitted for publication in a refereed journal. The journal that we have in mind is Agriculture, Ecosystems & Environment. In addition to this, we have extended this work to develop a new metric for investigating anthropogenic influences on water quality, the random runs value. This differs from the SCI in that, with the random runs value, the level of taxonomic identification is known.
Impacts
We expect this work to stimulate grass roots organizations that perform routine biomonitoring to investigate either adapting or creating their own sequential sampling programs.
Publications
- No publications reported this period
|
Progress 01/01/04 to 12/31/04
Outputs
In 2004, we investigated how selected binomial sequential sampling models could potentially influence the outcome of the sequential comparison index, or SCI, and attempted to determine how the estimate of the SCI is influenced by two levels of taxonomic expertise. For each of the selected binomial models, both non-technical as well as highly technically trained individuals sorted through samples of benthic macroinvertebrates to calculate the SCI. Non-technically trained individuals were given some basic instructions such as like-appearing organisms are to be considered as the same run, even though they are of different size while non-like-appearing organisms indicate the beginning of different runs. A repeated measures analysis of variance test indicated that, indeed, the non-technically trained individuals estimated the same SCI values as did the highly technically-trained individuals (P > 0.05).
Impacts
We expect this work to stimulate grass roots organizations that perform routine biomonitoring to investigate either adapting or creating their own sequential sampling programs.
Publications
- No publications reported this period
|
Progress 01/01/02 to 12/31/02
Outputs
In 2002, we were selected two binomial sequential sampling tools for further testing on 83 samples of benthic macroinvertebrates that were collected from various streams in and around southeast Wyoming (mostly). One of these tools was based on the Sequential Interval Procedure and made use of the binomial threshold of 0.2. The other was based on the Confidence Interval method and was based on the threshold of 0.8. Using these two tools, we wanted to see if they resulted in the same qualitative assessment of those 83 samples, as did processing the entire samples as well as taking a standard subsample of those samples. Results from comparing the outcomes of the two sequential sampling plans with the outcomes of the subsampling protocol as well as with the whole sampling protocol indicated that both sequential sampling plans required less time and fewer organisms be examined than did the subsampling or whole sampling protocols (time (min): F = 359.08, df = 3, 242, P <
0.0001, error mean square = 0.3996049) (sequential sampling plan that used a threshold of 0.2 = 1.86a+-0.095, sequential sampling plan that used a threshold of 0.8 = 14.52b+-1.449, subsampling protocol = 25.21c+-1.205, whole sampling protocol = 52.31d+-4.423) (number of organisms: F = 159.5, df = 3, 246, P < 0.0001, error mean square = 0.3565287) (sequential sampling plan that used a threshold of 0.2 = 10.72a+-0.416, sequential sampling plan that used a threshold of 0.8 = 70.31b+-6.410, subsampling protocol = 158.87c+-5.861, whole sampling protocol = 309.57d+-23.592). This result was expected, as Wald (1947) had reported a 50% savings in sampling effort was realized when using sequential sampling methods. Here, the savings in time and number or organisms examined ranged from 42.4% to 93.3% over the subsampling protocol and from 72.2% to 96.5% over the whole sampling protocol.
Impacts
We expect this work to stimulate grass roots organizations that perform routine biomonitoring to investigate either adapting or creating their own sequential sampling programs.
Publications
- Stitt, R. P. 2002. The effects of rapid laboratory assessment on the estimation of benthic macroinvertebrate diversity. M.S. Thesis, University of Wyoming, Laramie, Wyoming 82 pp.
|
|