Geology ReportsSearch

Geology topics

Clayton J. Williams

Publications and source records attributed to Clayton J. Williams.

3 recordsLinked to original sources

Estimated reductions in phosphorus loads from removal of leaf litter in the Lake Champlain drainage area, Vermont

Excess nutrient loading and other factors are driving eutrophication and other negative effects on water-quality conditions in Lake Champlain and other receiving waters in Vermont. Two common best management practices were evaluated to determine how these practices can be optimized by targeting maintenance and operation to align better with seasonally driven needs, specifically to help municipalities remove a greater proportion of seasonal leaves and organic debris, reduce nutrient loading, and achieve water-quality goals. To characterize solid materials typically removed by the municipal BMPs of catch-basin (CB) cleaning and street cleaning (SC), subsamples of CB and SC materials were collected each month from nine participating municipalities in central and northwestern Vermont between September 2017 and November 2018. Monthly and seasonal composites of CB and SC samples were created from the subsamples of available materials from all municipalities. Samples were analyzed for concentrations of total organic carbon, total Kjeldahl nitrogen, and total phosphorus (P), and separated into three particle-size fractions. Distribution of particle-size fractions was similar between CB and SC as both practices generally collect the coarser fraction of solid materials (greater than 125 micrometers in diameter). In the fall, however, the range of the coarser fraction of materials increased. This is attributed to the ability of SC to collect leaves and other light organic materials that commonly pass through a CB system designed to trap heavier materials. Total organic carbon, total Kjeldahl nitrogen, and total P concentrations were highest in the catch-basin samples in the fall of 2017, and concentrations in the SC samples were highest in the fall of 2018. The collection of fewer samples in 2017 may account for some of the variability between fall 2017 and fall 2018 results. A subset of SC samples collected from piles representing specific street-cleaning routes in September and November 2018 were also analyzed. Materials collected in November were dominated by leaves, and the concentrations of the analyzed species of carbon, nitrogen, and phosphorus in some samples were more than double those in samples collected on the same street-cleaning routes in September. The Vermont Department of Environmental Conservation and the University of Vermont developed estimates of load-reduction credits for CB and SC practices based on a policy developed by the Wisconsin Department of Natural Resources that determined the potential for credits associated with leaf-removal activities. This process also considered BMPs that were initiated during the U.S. Environmental Protection Agency’s Lake Champlain Basin Total Maximum Daily Load monitoring period (2000 to 2009) and adapted the Wisconsin Department of Natural Resources policies to apply to existing SC routes in the cooperating Vermont municipalities that possessed at least 17 percent tree cover. This exercise demonstrated that applying the Wisconsin Department of Natural Resources policy to existing street-cleaning routes possessing 17 percent or more tree cover would result in reductions in total P loads up to 65 percent of mandated target reductions, and about a 25 percent reduction on average. Continuous simulations of stormwater runoff volume, and of loads of suspended sediments and total P, also were created for Englesby Brook Basin, an urbanized basin in Burlington and South Burlington that drains to Lake Champlain. Although the basin is more developed than the average of the nine cooperating municipalities, streamflow and P loading data collected by the U.S. Geological Survey were available to evaluate model performance. Simulations based on a year of average climatic conditions projected potential small reductions in total P of 0.08 to 0.10 percent as a result of CB cleaning and SC practices. Simulated weekly SC practices, however, reduced street-solid loads by as much as 7 percent. When the proportion of total P seen in fall SC materials collected in Vermont was applied to these simulated street-solid loads, estimated reductions of total P were about 29 percent. The combination of analytical results, estimated load-reduction credits, and simulated reductions indicate that targeted increases of SC activities to reduce leaf loading in the fall have the potential to reduce loading to receiving waters and could help regulated communities meet their water-quality goals.

Vermont

Human activities cause distinct dissolved organic matter composition across freshwater ecosystems

Dissolved organic matter (DOM) composition in freshwater ecosystems is influenced by interactions between physical, chemical, and biological processes that are controlled, at one level, by watershed landscape, hydrology, and their connections. Against this environmental template, humans may strongly influence DOM composition. Yet, we lack a comprehensive understanding of DOM composition variation across freshwater ecosystems differentially affected by human activity. Using optical properties, we described DOM variation across five ecosystem groups of the Laurentian Great Lakes Region: large lakes, Kawartha Lakes, Experimental Lakes Area, urban stormwater ponds, and rivers (n = 184 sites). We determined how between ecosystem variation in DOM composition related to watershed size, land use and cover, water quality measures (conductivity, dissolved organic carbon (DOC), nutrient concentration, chlorophyll a ), and human population density. The five freshwater ecosystem groups had distinctive DOM composition from each other. These significant differences were not explained completely through differences in watershed size nor spatial autocorrelation. Instead, multivariate partial least squares regression showed that DOM composition was related to differences in human impact across freshwater ecosystems. In particular, urban/developed watersheds with higher human population densities had a unique DOM composition with a clear anthropogenic influence that was distinct from DOM composition in natural land cover and/or agricultural watersheds. This nonagricultural, human developed impact on aquatic DOM was most evident through increased levels of a microbial, humic-like parallel factor analysis component (C6). Lotic and lentic ecosystems with low human population densities had DOM compositions more typical of clear water to humic-rich freshwater ecosystems but C6 was only present at trace to background levels. Consequently, humans are strongly altering the quality of DOM in waters nearby or flowing through highly populated areas, which may alter carbon cycles in anthropogenically disturbed ecosystems at broad scales.

Global Change Biology

Relationships between land cover and dissolved organic matter change along the river to lake transition

Dissolved organic matter (DOM) influences the physical, chemical, and biological properties of aquatic ecosystems. We hypothesized that controls over spatial variation in DOM quantity and composition (measured with DOM optical properties) differ based on the source of DOM to aquatic ecosystems. DOM quantity and composition should be better predicted by land cover in aquatic habitats with allochthonous DOM and related more strongly to nutrients in aquatic habitats with autochthonous DOM. Three habitat types [rivers (R), rivermouths (RM), and the nearshore zone (L)] associated with 23 tributaries of the Laurentian Great Lakes were sampled to test this prediction. Evidence from optical indices suggests that DOM in these habitats generally ranged from allochthonous (R sites) to a mix of allochthonous-like and autochthonous-like (L sites). Contrary to expectations, DOM properties such as the fluorescence index, humification index, and spectral slope ratio were only weakly related to land cover or nutrient data (Bayesian R 2 values were indistinguishable from zero). Strongly supported models in all habitat types linked DOM quantity (that is, dissolved organic carbon concentration [DOC]) to both land cover and nutrients (Bayesian R 2 values ranging from 0.55 to 0.72). Strongly supported models predicting DOC changed with habitat type: The most important predictor in R sites was wetlands whereas the most important predictor at L sites was croplands. These results suggest that as the DOM pool becomes more autochthonous-like, croplands become a more important driver of spatial variation in DOC and wetlands become less important.

Great Lakes