You are here:
Spatial Patterns Study for Sediments from Lake Michigan
Citation:
XIA, X. AND D. H. MILLER. Spatial Patterns Study for Sediments from Lake Michigan. Presented at 2009 Annual IAGLR Conference, Toledo, OH, May 18 - 22, 2009.
Impact/Purpose:
Results of this study can be utilized for sediment assessments and modeling applications beyond Lake Michigan.
Description:
Accurately understanding the distribution of sediment measurements within large water bodies such as Lake Michigan is critical for modeling and understanding of carbon, nitrogen, silica and phosphorus dynamics. Several water quality models have been formulated and applied to the Great Lakes to investigate the fate and transport of nutrients and other constituents, as well as plankton dynamics. As part of the Lake Michigan Mass Balance Study, sampling of sediment measurements in Lake Michigan for nutrients which include data for phosphorus (total phosphorus, sodium hydroxide extractable phosphorus), biogenic silica, total organic nitrogen and organic carbon was conducted by the United States Environmental Protection Agency for the period of 1994 - 1996. In our modeling practices, data often need to be populated to cover a high resolution grid of 5 km by 5 km of Lake Michigan. Estimation algorithms which take into account and incorporate spatial patterns associated with data sets will generate better representative prediction in practice. However, spatial data analysis is known to be a difficult and time consuming process. Spatial statistics, in conjunction with descriptive statistics, were used to investigate and justify the spatial patterns associated with the sediment data sets. Our quantified results suggest that there is a strong correlation between the distribution of sediment parameters and the water depths where the samples were collected. Furthermore, the distribution of sediment nutrient and carbon measurements within Lake Michigan is observed to vary considerably from one area to another. We have explored a variety of approaches to incorporate the spatial patterns into data applications in order to get better statistical references. Two of the considerations in data applications were data stratification and 3D interpolations where knowledge of samples depths can be used to estimate variogram and covariance functions. Test results have shown improvements over the traditional estimation procedures conducted without using spatial patterns. Results of this study can be utilized for sediment assessments and modeling applications beyond Lake Michigan.