Hydrologic landscape regions (HLRs) in the United States were delineated by using geographic information system (GIS) tools and statistical methods including principal components and cluster analyses. The GIS and statistical analyses were applied to land-surface form, geologic texture (permeability of the soil and bedrock), and climate variables that describe the physical and climatic setting of 43,931 small (roughly 200 square kilometers) watersheds in the United States. The analyses then grouped the watersheds into 20 noncontiguous regions (the HLRs) on the basis of similarities in land-surface form, geologic texture, and climate characteristics. This hydrologic landscape regions dataset contains for each of the 49,931 watersheds the (1) watershed identification number, (2) land-surface form, geologic texture, and climate characteristics for each watershed, and (3) hydrologic landscape region number for each watershed.

The purpose of defining hydrologic landscape regions was to group watersheds in the United States according to their similarity in landscape and climate characteristics. The selected landscape and climate characteristics represent factors assumed to affect hydrologic processes in the environment. The hydrologic landscape regions grid was used to help design the sampling network in the U.S. Geological Survey's National Water-Quality Assessment (NAWQA) Program. For use with the GEOSTAC database, this data set has been compiled in order to simplify pesticide risk assessments and provide a common data set for vested interests.

The U.S. Geological Survey (USGS) initiated the National Water-Quality Assessment (NAWQA) Program in 1991 to accomplish two primary objectives. The first objective was to assess the status and trends in the quality of the Nation's ground-water and surface-water resources, and the second objective was to link the status and trends with an understanding of the natural and human factors that affect water quality (Gilliom and others, 1995). The original NAWQA implementation plan was based on assessing water quality in 60 study areas located throughout the United States. The extensive geographic coverage of the 60 study areas was considered adequate to sample the diversity of hydrologic settings throughout the Nation. The characteristics of hydrologic-cycle components, such as precipitation, evapotranspiration, infiltration, ground-water flow, and overland flow, were used to define the hydrologic setting of each study area. Adequate coverage of the diversity in hydrologic settings is crucial in understanding how natural and human factors affect water quality. The second decade of the NAWQA Program is well underway. One significant constraint on refining the design of the program is budget; the current funding outlook for 2002-11 can support only about 40 of the original 60 study areas. The need to maintain a wide and representative sample of the hydrologic settings throughout the United States remains imperative. The challenge, then, was to carefully select study areas and specific data-collection sites that represent the entire range of hydrologic settings. Selecting a group of study areas to represent the range of hydrologic settings in the United States required a digital map of the factors expected to affect these hydrologic settings. The ideal map needed was one that divided the United States into regions, where areas in the same region have similar hydrologic conditions and areas in different regions have different hydrologic conditions. A region would not necessarily be spatially contiguous; areas similar to each other can be located in different parts of the United States. The grouping of areas into regions on the basis of similarities in landscape and climate characteristics is referred to in this documentation as a "regional framework." Regional frameworks related to hydrologic characteristics have been used in many water-resource and water-quality applications. Most significantly, ecoregions (Omernik, 1987) have been used to define areas where conditions affecting stream chemistry are expected to be similar (Omernik and Griffith, 1991). Ecoregions, however, were not specifically designed to identify areas throughout the United States with similar hydrologic characteristics. Rather, ecoregions were developed to identify patterns in biotic and abiotic factors thought to generally affect ecological processes at a relatively broad scale. For the purpose of selecting areas to be studied in the second decade of NAWQA, a map of hydrologic landscape regions (HLRs) based on the hydrologic landscapes concept of Winter (2001) was produced. The hydrologic landscapes concept reflects fundamental hydrologic processes that are known to affect water quality and other environmental characteristics. A map of HLRs was generated using readily available spatial data layers, geographic information system (GIS) tools, and statistical analyses to produce a consistent and reproducible hydrologic characterization of the United States. The regions in this dataset were derived specifically to select study areas for the NAWQA Program. If the application were for a different purpose, covered a different spatial extent, or was based on a different dataset, then the final product (that is, the regions) would be different. The hydrologic landscapes concept The characteristics of the Earth and its climate that affect the location, movement, and chemistry of water are extremely complex. With respect to the movement of water, however, many seemingly diverse landscapes have some features in common. It is these commonalties that need to be identified in developing a conceptual framework. Only by evaluating landscapes from a common conceptual hydrologic framework can processes common to some or all landscapes be distinguished from processes unique to particular landscapes. The concept of hydrologic landscapes is based on the idea that a single physical feature is the basic building block of all landscapes. This feature is termed a fundamental hydrologic landscape unit and is defined as an upland adjacent to a lowland separated by a valley side. The hydrologic system of a fundamental hydrologic landscape unit consists of (1) the movement of surface water, which is controlled by the slopes and permeability of the landscape; (2) the movement of ground water, which is controlled by the hydraulic characteristics of the geologic framework; and (3) atmospheric-water exchange, which is controlled by climate (Winter, 2001). All hydrologic landscapes can be conceived of as variations and multiples of fundamental hydrologic landscape units, and these can be used to define general landscape types that describe major physiographic features of the Earth. Some examples are: (1) a landscape consisting of narrow lowlands and uplands separated by high and steep valley sides, characteristic of mountainous terrain; (2) a landscape with wide lowlands separated from much narrower uplands by steep valley sides, characteristic of basin-and-range physiography and basins of interior drainages; and (3) a landscape having narrow lowlands separated from very broad uplands by valley sides of various slopes and heights, characteristic of plateaus and high plains. The movement of water over the surface and through the subsurface of generalized landscapes is controlled by common physical principles regardless of the geographic location of the landscapes. For example, in a landscape with low permeability soils, surface runoff will be extensive, and recharge to ground water will be limited. In contrast, in a landscape with highly permeable soils and steep slopes, surface runoff will be limited, and ground-water recharge will be significant. Moreover, in landscapes that have a shallow water table, transpiration directly from ground water may have a substantial effect on the volume of ground water and on the movement of ground water to and from surface water. ----------------------------------------------------------------------- Cited references Gilliom, R.G., Alley, W.M., and Gurtz, M.E., 1995, Design of the National Water-Quality Assessment Program--occurrence and distribution of water-quality conditions: U.S. Geological Survey Circular 1112, 33 p. Omernik, J.M., 1987, Ecoregions of the conterminous United States: Annals of the Association of American Geographers, v. 77, p. 118-125. Omernik, J.M., and Griffith, G.E., 1991, Ecological regions versus hydrologic units--frameworks for managing water quality: Journal of Soil and Water Conservation, v. 46, p. 334-340. Winter, T.C., 2001, The concept of hydrologic landscapes: Journal of the American Water Resources Association, v. 37, p. 335-349.

The HLRs derived in this study are specific to several choices made in the analysis: (1) the particular set of variables used, (2) details in the statistical analyses, and (3) details in the GIS analyses. Using a different set of watershed characteristics would have affected the derived HLR map. In addition, averaging the variables over smaller areas would have produced a map with finer spatial detail. Changing any of these factors likely would lead to other regional maps that could be equally valid and useful. The methods used to define HLRs are expected to be sensitive to the spatial scale of the analysis. In the study described herein, the spatial extent of the analysis covered all 50 States; this was the appropriate spatial scale for the purpose of identifying HLRs to help design a national water-quality assessment. Satisfying a different objective may require a different spatial scale of analysis and might result in a different set of regions. The concept of hydrologic landscapes proposed by Winter (2001) represents the natural landscape and climate factors expected to affect hydrologic processes. There are important water-resources management activities, such as reservoir and canal construction, irrigation, and subsurface drainage, that have a significant effect on hydrologic processes. These water-resources management activities were not included as factors in defining the HLRs. Cited references: Winter, T.C., 2001, The concept of hydrologic landscapes: Journal of the American Water Resources Association, v. 37, p. 335-349.

Process_Description:

Delineation of hydrologic landscape regions

The overall approach for delineation of HLRs for the United States involved three steps:
(1) Delineate a set of watersheds that covers all 50 States.
(2) Determine metrics to quantify the land-surface form, geologic texture (herein defined as the permeability of the soil and bedrock), and climate characteristics that define hydrologic landscapes, and then average the metrics for each watershed.
(3) Use cluster analysis to assign the watersheds to groups according to their similarity in land-surface form, geologic texture, and climate characteristics. Each group of similar watersheds comprises an HLR.

ARC/INFO GIS (Environmental Systems Research Institute, Inc., 1991) was used in the first two steps just listed. (The use of firm, trade, and brand names in this document is for identification purposes only and does not constitute endorsement by the U.S. Geological Survey.) All the GIS data were analyzed at 1-km resolution in the Lambert Azimuthal Equal Area projection system (central meridian, -100 deg; latitude of origin, 45 deg; sphere of influence radius, 6,370,997 m).

Delineation of watersheds

A set of small watersheds (each about 200 square kilometers in area) completely covering the 50 States was derived from 1-kilometer resolution digital elevation model (DEM) data (Verdin and Greenlee, 1996). This watershed size was selected because it is compatible with the spatial resolution of nationally available datasets on land-surface form, geologic texture, and climate characteristics (see the next section). This watershed size is large enough to include lowlands and uplands, but small enough to provide spatial detail among watersheds.

First, a synthetic stream network was extracted from the DEM by specifying a minimum drainage area of 100 square kilometers to initiate a stream channel. Each stream segment in the network was assigned a unique identifier value, and the entire set of stream segments then was used to delineate the watersheds from the DEM. The result was a set of 43,931 watersheds with an average area of 212 square kilometers. It is important to note that DEM-based watershed delineation is impossible in completely flat areas. Therefore, each spatially contiguous flat area (that is, adjacent DEM grid cells with the same elevation) became a single, possibly large, watershed.

Quantification of land-surface form, geologic texture, and climate characteristics

The critical features used to describe Winter's hydrologic landscapes are: (1) land-surface form, to quantify the effects of gravity on the movement of water through the landscape; (2) geologic texture, to estimate permeability of soil and bedrock materials that affect surface runoff, infiltration, and ground-water flow; and (3) climate characteristics, to approximate water available to surface- and ground-water systems.

Land-surface form is described by terrain characteristics that are computed from 1-km resolution DEM data for the small watersheds. The characteristics are: relief (maximum elevation minus minimum elevation in the watershed); total percentage of flatland (areas with less than 1-percent slope) in the watershed; the percentage of flatland located in upland areas of the watershed (flat areas with elevation greater than the midpoint elevation); and the percentage of flatland in lowland areas of the watershed (flat areas with elevation less than or equal to the midpoint elevation). These terrain characteristics are similar to those in Hammond (1964) and presented in the USGS National Atlas of the United States "Classes of Land-Surface Form" map (U.S. Geological Survey, 2001). The land-surface form characteristics were computed for each of the delineated watersheds.

Geologic texture is described by two measures of permeability--one for soil and one for bedrock material. Soil permeability is estimated as the percentage of sand in the soil listed in the U.S. Department of Agriculture's STATSGO database (U.S. Department of Agriculture, 1994). The percentage of sand in the soil is directly related to permeability; that is, permeability increases as percentage of sand in the soil increases.

Bedrock permeability is quantified by assigning permeability classes to the general lithologic groups (table 1) in the USGS National Atlas map of principal aquifer groups (U.S. Geological Survey, 2001). These bedrock texture divisions are consistent with the primary lithologies used to develop regions for the USGS Ground-Water Resources Program (U.S. Geological Survey, 1998). Some of the bedrock texture groups have permeability values ranging over several orders of magnitude, such as basalt and volcanic rocks, but "typical" permeability is the basis for the relative permeability class.
--------------------------------------------------------
Table 1. Lithologic groups of principal aquifers and bedrock permeability classes [1 is the lowest permeability and 7 is the highest]

Lithologic group (bedrock permeability class)

Not a principal aquifer (1)
Sandstone (2)
Semiconsolidated sand (3)
Basalt and other volcanic rocks (4)
Sandstone and carbonate rocks (5)
Unconsolidated sand and gravel (6)
Carbonate rock (7)
-----------------------------------------------------------

The digital maps representing geologic texture were intersected with the delineated watersheds to compute average percentage of sand in the soil and average bedrock permeability class for each of the 43,931 watersheds.

Climate characteristics are described by mean annual precipitation minus potential evapotranspiration (PET). PET was estimated from mean monthly temperature and latitude using the Hamon equation (Hamon, 1961). Mean monthly temperature and precipitation data from first-order meteorological station data for 1961-90 (Owensby and Ezell, 1992) were used to compute the mean annual precipitation and PET estimates. The station data were interpolated to a 1-km resolution grid using the ARC/INFO inverse-distance weighting method. The mean value of precipitation-PET was computed for each of the 43,931 watersheds by intersecting the digital precipitation-PET data with the delineated watersheds.

Identification of hydrologic landscape regions by grouping the watersheds

The land-surface form, geologic texture, and climate characteristics were used to assign the watersheds to similar groups that define the HLRs for the 50 States. First, a principal components analysis of the land-surface form, geologic texture, and climate characteristics for the 43,931 watersheds was used to remove co-dependence among the variables, and then a Fortran-language clustering program was used to assign each watershed to 1 of 20 groups. The 20 groups were selected on the basis of low within-cluster variance and good separation among the clusters. The hierarchical method in the clustering program uses a minimum variance criterion and the nearest neighbor chain algorithm (Murtagh, 1985).

The GIS and statistical analyses resulted in a grid of 20 HLRs. Statistics (such as mean and standard deviation) of the landscape characteristics can be computed using zonal statistics functions in a GIS. The numbers and associated colors assigned to the HLRs were chosen according to the overall similarity in hydrologic-landscape characteristics (land-surface form, geologic texture, and climate characteristics) among regions. For example, HLR 1 and HLR 2 are similar in their average hydrologic-landscape characteristics except for bedrock permeability class. In contrast, HLR 1 and HLR 20 are very different in all characteristics except precipitation-PET.

An important attribute of HLRs is that the same region can occur in different parts of the United States. A notable example of this feature is HLR 1, which is located in the Southeast, the western Great Lakes area, and the Sand Hills of Nebraska. This indicates that the hydrologic-landscape characteristics of watersheds in these geographically disparate sections of the United States are similar. The Sand Hills of Nebraska has lower precipitation-PET than the Southeast and western Great Lakes area. However, the overall similarity in the three regions is greater than their differences.

The particular combination of land-surface form, geologic texture, and climate characteristics can be used to qualitatively describe a given hydrologic landscape (Table 2).

--------------------------------------------------------------------
Table 2. Hydrologic landscape region (HLR) descriptions

(HLR number) Description

(1) Subhumid plains with permeable soils and bedrock
(2) Humid plains with permeable soils and bedrock
(3) Subhumid plains with impermeable soils and permeable bedrock
(4) Humid plains with permeable soils and bedrock
(5) Arid plains with permeable soils and bedrock
(6) Subhumid plains with impermeable soils and bedrock
(7) Humid plains with permeable soils and impermeable bedrock
(8) Semiarid plains with impermeable soils and bedrock
(9) Humid plateaus with impermeable soils and permeable bedrock
(10) Arid plateaus with impermeable soils and permeable bedrock
(11) Humid plateaus with impermeable soils and bedrock
(12) Semiarid plateaus with permeable soils and impermeable bedrock
(13) Semiarid plateaus with impermeable soils and bedrock
(14) Arid playas with permeable soils and bedrock
(15) Semiarid mountains with impermeable soils and permeable bedrock
(16) Humid mountains with permeable soils and impermeable bedrock
(17) Semiarid mountains with impermeable soils and bedrock
(18) Semiarid mountains with permeable soils and impermeable bedrock
(19) Very humid mountains with permeable soils and impermeable bedrock
(20) Humid mountains with permeable soils and impermeable bedrock
-----------------------------------------------------------------------

Cited references

Environmental Systems Research Institute, Inc., 1991, Understanding GIS--the ARC/INFO method: Redlands, California, various pagination.

Hammond, E.H., 1964, Classes of land-surface form in the 48 states, USA: Annals of the Association of American Geographers, v. 54, no. 1964, map supplement no. 4, 1:5,000,000 scale.

Hamon, W.R., 1961, Estimating potential evapotranspiration: Journal of the Hydraulics Division, Proceedings of the American Society of Civil Engineers, v. 87, p. 107-120.

Murtagh, F., 1985, Multidimensional clustering algorithms: Vienna, Physica-Verlag.

Owensby, J.R., and Ezell, D.S., 1992, Climatography of the United States, No. 81--monthly station normals of temperature, precipitation, and heating and cooling degree days, 1961-90: Ashville, North Carolina, U.S. Department of Commerce, National Oceanic and Atmospheric Administration, National Climatic Data Center.

U.S. Department of Agriculture, 1994, State Soil Geographic STATSGO data base--data use information: Miscellaneous Publication Number 1492, 35 p.

U.S. Geological Survey, 1998, Strategic directions for the U.S. Geological Survey Ground-Water Resources Program--a report to Congress, November 30, 1998: Information available on the World Wide Web, accessed March 24, 2003, at URL http://water.usgs.gov/ogw/gwrp/stratdir/stratdir.html

U.S. Geological Survey, 2001, National atlas of the United States maps: U.S. Geological Survey Fact Sheet 086-01, information available on the World Wide Web, accessed March 24, 2003, at URL http://mac.usgs.gov/mac/isb/pubs/factsheets/fs08601.html

Verdin, K.L., and Greenlee, S.K., 1996, Development of continental scale digital elevation models and extraction of hydrographic features, In Proceedings, Third International Conference/Workshop on Integrating GIS and Environmental Modeling, Santa Fe, New Mexico, January 21-26, 1996: Santa Barbara, California, National Center for Geographic Information and Analysis.

The attributes give climate, soils, terrain, and geologic characteristics for each of the defined watersheds. The hydrologic landscape regions numbers indicate similar watersheds; watersheds with the same hydrologic landscape region number are most similar to each other. Watersheds with very different HLR numbers (for example, 1 and 20) are very different from each other in their landscape characteristics.

Although these data have been used by the U.S. Geological Survey, U.S. Department of the Interior, no warranty expressed or implied is made by the U.S. Geological Survey as to the accuracy of the data. The act of distribution shall not constitute any such warranty, and no responsibility is assumed by the U.S. Geological Survey in the use of this data, software, or related materials.

hlrus

Metadata: