Search

We present a parametric model for estimating the areas within watersheds whose land use best predicts indicators of stream ecological condition. We regress a stream response variable on the distance-weighted proportion of watershed area that has a specific land use, such as agriculture. Distance weighting functions model the declining influence of landscape elements as a function of their flowpath distances, first to the stream channel (to-stream distance), and then down the channel to the location at which stream condition was sampled (in-stream distance). Model parameters specify different distance scales over which to-stream and in-stream influences decline. As an example, we predict an index of biotic integrity (IBI) for the fish communities in 50 small streams of the Willamette Basin of Oregon, USA, from distance-weighted proportions of agricultural or urban land use in their watersheds. The weighting functions of best-fitting models (R 2 = 0.57) represent landscape influence on IBI as extending upstream tens of kilometers along the stream channel network, while declining nearly to zero beyond a distance of 30 m from the channel. Our example shows how parametric distance weighting can identify the distance scales, and hence the approximate areas within watersheds, for which land use is most strongly associated with a stream response variable. In addition, distance-weighting parameters offer a simple and direct language for comparing the scales of landscape influence on streams across different land uses and stream ecosystem components.

Conventional soil texture classification systems use different definitions of particle-size distributions (PSDs). For example, sand in the International Soil Science Society (ISSS) system equals the combined separate limits of coarse silt and sand in the USDA system. Because relationships between texture and other soil properties are affected by these differences, the ability to merge survey data in environmental studies is limited. Previous research calculated two PSD statistics, namely the geometric mean particle diameter (dg) and its standard deviation (σg), which do not depend on separate limits. We expanded the development of the PSD statistics dg and σg to compare the USDA and ISSS systems, develop relationships with soil properties, include rock fragments, and simplify the USDA texture classification to facilitate the use of soil survey data in environmental research. We found that (i) for equal clay and sand fractions, the texture of a soil sample as described by the USDA system has larger dg and σg values than in ISSS; (ii) for equal clay and sand fractions, soil samples have larger values of cation-exchange capacity (CEC) in the ISSS than in the USDA system; (iii) small differences between some of the traditional 12 USDA classes are reflected in the dg and σg values for samples containing rocks, thereby presenting a rationale for simplification; and (iv) with this rationale, the 12 USDA classes were aggregated into five classes.

The Surface Waters component of the Environmental Monitoring and Assessment Program (EMAPSW) was developed by the USEPA to evaluate the extent and condition of lakes and streams over national and regional scales. Realistically, chemical or physical water properties (WPs) such as acidity or turbidity can be field-sampled for only a small portion of all lakes and streams. However, soil characteristics (SCs) affect WPs and broad-scale soil survey data have become available in the State Soil Geographic Data Base (STATSGO). We developed models relating observed WPs to SCs to extrapolate the sampled WPs to a region, potentially reducing extensive monitoring needs. Our study region consisted of 13 northeastern and Mid-Atlantic states and contained 882 STATSGO soil map units. We used map units as the spatial component of WP analysis. The WPs were sampled in 721 randomly selected EMAPSW study sites. The watersheds of these sites represent 7.1% of the region's total area and spatially intersect 400 of its soil map units. Each intersected map unit was assigned the weighted average WPs from the corresponding watersheds. Conditional expectation models were used to extrapolate sampled WPs to 882 map units. The relative standard errors ranged from low for pH (0.8%), intermediate for total P (12.1%), and very high for chloride (54.8%). The high extrapolation errors indicate outlier conditions from natural, non-soil, or anthropogenic sources.

Soils support ecosystem functions such as plant growth and water quality because of certain physical, chemical, and biological properties. These properties have been studied at different spatial scales, including point scales to satisfy basic research needs, and regional scales to satisfy monitoring needs. Recently, soil property data for the entire USA have become available in the State Soil Geographic Data Base (STATSGO), which is appropriate for regional-scale research. We analyzed and created models of STATSGO data in this study to serve as a research tool, for example, for linking the soil to regional water quality monitoring data in our companion paper. Map units in STATSGO define geographic land areas by soil characteristics (SCs) of similar soil series. We selected 27 SCs that influenced water properties (in varying degrees), aggregated the layer and component SCs to map unit SCs, and used SCs to calculate relationships among map units. The relationships were defined by equations of conditional mean for the q * SC (SCq), while using the remaining 26 SCs as predictors. The relative standard errors for 22 of the 27 SCs were less than 10%, and less than 22% for the remaining five. We conclude that spatial extrapolation of SCs is feasible and the procedures are a first step toward extrapolating information across a region using SC-water property relationships. Although our procedure is for regional scale monitoring, it is also applicable to finer spatial scales commensurate with available soil data.

We compared different sources of land use/land cover data using the percentages of forest, agriculture, urban, and mined lands in approximately 400 watersheds of the U.S. Mid-Atlantic region. The primary data sources were digital Land Use/Land Cover maps (LULC) from the U.S. Geological Survey (USGS) and data derived from Landsat Thematic Mapper (TM) imagery obtained by the Multi-Resolution Land Characteristics (MRLC) Consortium and interpreted at the Earth Resources Observation Systems (EROS) Data Center (version 2). The latter is referred to as MRLC-TM. We also used aerial photographs (AP), topographic maps, field comments, and qualitative watershed assessments as ancillary information. The LULC and MRLC-TM data differed in age, source material, classifications, format (vector vs. raster), and resolution. However, Pearson correlation coefficients between LULC and MRLC-TM were high for broad (composite) forest, agriculture, and urban categories (0.95, 0.96, and 0.92, respectively). Correspondence varied b...

Researchers have defined and mapped ecological regions of the United States based on similar patterns of ecosystems such as deserts, forests, and croplands. These studies are useful in regional research, monitoring, and environmental management because data can be more readily extrapolated within the same ecoregion and to regions with similar characteristics. The description of ecoregions is largely holistic and qualitative. Conversely, quantitative information for soil are abundant and soil is an important ecosystem component related to many ecoregion properties. We used the nationwide State Soil Geographic database (STATSGO) to describe the soils of 84 Level III ecoregions in the United States. Among the 24 soil characteristics studied were texture, rock fragments, available water capacity, bulk density, and organic matter content. For each ecoregion we developed ranks to describe (i) its similarity to the U.S. average soil characteristics, (ii) the accuracy of predicting those characteristics, (iii) how well the soil map unit boundaries fit within ecoregion boundaries, (iv) the spatial relationship of soils across neighboring ecoregion boundaries, and (v) the homogeneity of texture-rock patterns. We present a national map of soil texture and rock fragments and five soil ranks for each ecoregion, and examine relationships between soils and other ecological components for selected ecoregions. Because soils relate to other ecosystem components such as vegetation, geology, and land use, the soil ranks complement and enrich the qualitative ecoregion descriptions. Similar analyses of physical or biological components of ecoregions will expand the understanding of the ecosystem patterns.

In order to investigate the relationship between stream chemistry and watershed land cover at the regional scale, we analyzed data from 368 wadeable streams sampled in the mid-Atlantic region of the U.S. during spring 1993–1994. Study sites were selected using a probability sample and the digitized version of the 1:100,000 scale USGS map stream network as the sample population. Both classified Thematic Mapper (TM) and USGS Land Use/Land Cover (LULC) data were used to quantify land cover in the study watersheds. On average, the most common land cover was forest (77%) followed by agriculture (20%), and urban (1%). Multiple regression analysis showed that concentrations of Cl−, nutrients, acid neutralization capacity, and base cations were the analytes most strongly related to watershed land cover. Despite large differences in resolution and age of the TM and LULC data sources, similar results were obtained with the two sources. Using a greater number of land cover subclasses did not greatly improve the land cover-chemistry relationships. Ecoregions with predominantly forested land cover had weaker relationships than ecoregions with more agricultural and/or urban land cover. In studies or databases without land cover information, Cl− concentration is a good surrogate indicator for general human disturbance in the watershed.

Topographic maps are commonly used to define populations of lakes in regional surveys of surface water quality. To illustrate the effect of different maps on that process, we compared the lakes represented on the 1:250,000-scale maps used for the Northeast Region of the Eastern Lake Survey—Phase I (ELS-I) to the lakes on a sample of large-scale maps (1:24,000 or 1:62,500). Lake areas at or near the lower limit of representation delimited “smallest-lake” values for the compared 1:250,000-scale maps. The regional median for these values was 4.5 hectares (ha) and ranged from 0.6 to 24.8 ha. Lake representation is influenced by cartographic limitations such as map scale, age, and complexity as well as the inherent variability of waterbodies (e.g., water level fluctuations or the creation of reservoirs, beaver impoundments, and oxbows). The total number of lakes on large-scale maps increased markedly as lake area decreased. Approximately 15,700 of the estimated 29,000 lakes in the EPA's Northeast Region were 1 to 4 ha in area. Because maps affect the size distribution of lakes included in a regional survey and because lake areas are thought to modify lake chemistry, maps ultimately affect the estimates of regional surface water quality.