AUTOMATED EXTRACTION OF DRAINAGE NETWORK AND WATERSHED DATA FROM DIGITAL ELEVATION MODELS1

ABSTRACT: This paper discusses a computer program which extracts a number of watershed and drainage network properties directly from digital elevation models (DEM) to assist in the rapid parameterization of hydrologic runoff models. The program integrates new and established algorithms to address problems inherent in the analysis low-relief terrain from raster DEMs similar to those distributed by the U.S. Geological Survey for 7.5-minute quadrangles. The program delineates the drainage network from a DEM, and determines the Strahler order, total and direct drainage area, length, slope, and upstream and downstream coordinates of each channel link. It also identifies the subwatershed of each channel source and of the left and right bank of each channel link, and assigns a unique number to each network node. The node numbers are used to associate each subwatershed with the channel link to which it drains, and can be used to control flow routing in cascade hydrologic models. Program output includes tabular data and raster maps of the drainage network and subwatersheds. The raster maps are intended for import to a Geographical Information System where they can be registered to other data layers and used as templates to extract additional network and subwatershed information.     

STATISTICAL LOW FLOW ESTIMATION USING GIS ANALYSIS IN HUMID MONTANE REGIONS IN PUERTO RICO1

ABSTRACT: Statistical analysis of watershed parameters derived using a Geographical Information system (GIS) was done to develop equations for estimating the 7d–10yr, 30d–10yr, and 7d–2yr low flow for watersheds in humid montane regions of Puerto Rico. Digital elevation models and land use, geology, soils, and stream network coverages were used to evaluate 21 geomorphic, 10 stream channel, 9 relief, 7 geology, 4 climate, and 2 soil parameters for each watershed. To assess which parameters should be used for further investigation, a correlation analysis was used to determine the independence and collinearity among these parameters and their relationship with low flows. Multiple regression analyses using the selected parameters were then performed to develop the statistical models of low flows. The final models were selected in the basis of the Mallow Cp statistic, the adjusted R², the Press statistic, the degree of collinearity, and an analysis of the residuals. In the final models, drainage density, the ratio of length of tributaries to the length of the main channel, the percent of drainage area with northeast aspect, and the average weighted slope of the drainage were the most significant parameters. The final models had adjusted standard errors of 58.7 percent, 59.2 percent, and 48.6 percent for the 7d–10yr, 30d–10yr, and 7d–2yr low flows respectively. For comparison, the best model based on watershed parameters that can be easily measured without a GIS had an adjusted standard error of 82.8 percent.     

Topographic Effects on Soil Organic Carbon in Louisiana Watersheds

Terrestrial carbon storage is influenced by a number of environmental factors, among which topographic and geomorphological features are of special significance. This study was designed to examine the relationships of soil organic carbon (SOC) density to various terrain parameters and watershed characteristics across Louisiana, USA. A polygon data set of 484 watersheds and 12 river drainage basins for Louisiana was used to form the landscape units. SOC densities were calculated for each soil map unit using the State Soil Geographic (STATSGO) database. Average drainage densities and average slopes at watershed and basin scales were quantified with the 1:24 K Digital Elevation Models (DEM) data, and the Louisiana hydrographic water features. Correlation and regression analyses were performed to determine relationships among drainage density, slope, elevation, and SOC. The study found an average watershed drainage density of 1.6 km/km² and an average watershed slope of 2.9 degrees in Louisiana. The results revealed that SOC density at both watershed and basin scales was closely related to drainage density, slope, and elevation. SOC density was positively correlated with watershed drainage density, but negatively correlated with watershed slope gradient and elevation. Regression models were developed for predicting SOC density at watershed and basin scales, obtaining regression coefficients (r ²) ranging from 0.43 to 0.83. The study showed that estimation of SOC at watershed and drainage basin scales combining DEM data can be a feasible approach to improve the understanding of the relationships among SOC, topographic, and geomorphological features.     

A GEOGRAPHIC INFORMATION SYSTEM PROCEDURE TO QUANTIFY DRAINAGE-BASIN CHARACTERISTICS1

ABSTRACT: The Basin Characteristics System (BCS) has been developed to quantify characteristics of a drainage basin. The first of four main BCS processing steps creates four geographic information system (GIS) digital maps representing the drainage divide, the drainage network, elevation contours, and the basin length. The drainage divide and basin length are manually digitized from 1:250,000-scale topographic maps. The drainage network is extracted using GIS software from 1:100,000-scale digital line graph data. The elevation contours are generated using GIS software from 1:250,000-scale digital elevation model data. The second and third steps use software developed to assign attributes to specific features in three of the four digital maps and analyze the four maps to quantify 24 morphometric basin characteristics. The fourth step quantifies two climatic characteristics from digitized State maps of precipitation data. Compared to manual methods of measurement, the BCS provides a reduction in the time required to quantify the 26 basin characteristics. Comparison tests indicate the BCS measurements are not significantly different from manual topographic-map measurements for 11 of 12 primary drainage-basin characteristics. Tests indicate the BCS significantly underestimates basin slope. Comparison-measurement differences for basin slope, main channel slope, and basin relief appear to be due to limitations in the digital elevation model data.     

EXTRACTION AND UTILIZATION OF SPACE ACQUIRED PHYSIOGRAPHIC DATA FOR WATER RESOURCES DEVELOPMENT1

ABSTRACT: ERTS-1 satellite imagery has been evaluated as a means of providing useful watershed physiography information. From these data physiographic parameters such as drainage basin area and shape, drainage density, stream length and sinuosity, and the percentage of a watershed occupied by major land use types were obtained in three study areas. The study areas were: (1) Southwestern Wisconsin; (2) Eastern Colorado; and (3) portions of the Middle Atlantic States Using ERTS-1 imagery at 1:250,000 and 1:100,000 scales it was found that drainage basin area and shape and stream sinuosity were comparable (within 10%) in all study areas to physiographic measurements derived from conventional topographic maps at the same scales Land use information can be usefully extracted for watersheds as small as 30 mi²(78 km²) in area. Improved drainage network and density information is obtained from ERTS-1 imagery in dissected areas such as Southwestern Wisconsin, but in heavily vegetated areas (Middle Atlantic States) or areas with little physical relief (Eastern Colorado) low order streams are difficult to detect and the derived drainage densities are significantly smaller than those obtained from standard maps. It is concluded that ERTS-1 imagery can be employed to advantage in mean annual runoff prediction techniques and in providing or maintaining land use information used in the calibration and operation of watershed models.     

Evaluating the Slope‐Area Method to Accurately Identify Stream Channel Heads in Three Physiographic Regions

Estimation of stream channel heads is an important task since ephemeral channels play a significant role in the transport of sediment and materials to perennial streams. The slope‐area method utilizes digital elevation model (DEM) and related information to develop slope‐area threshold relationships used to estimate the position of channel heads in the watershed. A total of 162 stream channel heads were mapped across the three physiographic regions of Alabama, including the Southwestern Appalachians (51), Piedmont/Ridge and Valley (61), and Coastal Plains (51). Using Geographic Information System and DEM, the local slope and drainage area for each mapped channel head was calculated and region‐specific models were developed and evaluated. Results demonstrated the local slope and drainage area had an inverse and strong correlation in the Piedmont/Ridge and Valley region (r² = 0.71) and the Southwestern Appalachian region (r² = 0.61). Among three physiographic regions, the weakest correlation was observed in the Coastal Plain region (r² = 0.45). By comparing the locations of modeled channel heads to those located in the field, calculated reliability and sensitivity indices indicated model accuracy and reliance were weak to moderate. However, the slope‐area method helped define the upstream boundaries of a more detailed channel network than that derived from the 1:24,000‐scale National Hydrography Dataset, which is commonly used for planning and regulatory purposes.     

Development of a SWAT Patch for Better Estimation of Sediment Yield in Steep Sloping Watersheds1

Abstract: The watershed scale Soil and Water Assessment Tool (SWAT) model divides watersheds into smaller subwatersheds for simulation of rainfall‐runoff and sediment loading at the field level and routing through stream networks. Typically, the SWAT model first needs to be calibrated and validated for accurate estimation through adjustment of sensitive input parameters (i.e., Curve Number values, USLE P, slope and slope‐length, and so on). However, in some instances, SWAT‐simulated results are greatly affected by the watershed delineation and Digital Elevation Models (DEM) cell size. In this study, the SWAT ArcView GIS Patch II was developed for steep sloping watersheds, and its performance was evaluated for various threshold values and DEM cell size scenarios when delineating subwatersheds using the SWAT model. The SWAT ArcView GIS Patch II was developed using the ArcView GIS Avenue program and Spatial Analyst libraries. The SWAT ArcView GIS Patch II improves upon the SWAT ArcView GIS Patch I because it reflects the topographic factor in calculating the field slope‐length of Hydrologic Response Units in the SWAT model. The simulated sediment value for 321 subwatersheds (watershed delineation threshold value of 25 ha) is greater than that for 43 subwatersheds (watershed delineation threshold value of 200 ha) by 201% without applying the SWAT ArcView GIS Patch II. However, when the SWAT ArcView GIS Patch II was applied, the difference in simulated sediment yield decreases for the same scenario (i.e., difference in simulated sediment with 321 subwatersheds and 43 subwatersheds) was 12%. The simulated sediment value for DEM cell size of 50 m is greater than that for DEM cell size of 10 m by 19.8% without the SWAT ArcView GIS Patch II. However, the difference becomes smaller (3.4% difference) between 50 and 10 m with the SWAT ArcView GIS Patch II for the DEM scenarios. As shown in this study, the SWAT ArcView GIS Patch II can reduce differences in simulated sediment values for various watershed delineation and DEM cell size scenarios. Without the SWAT ArcView GIS Patch II, variations in the SWAT‐simulated results using various watershed delineation and DEM cell size scenarios could be greater than those from input parameter calibration. Thus, the results obtained in this study show that the SWAT ArcView GIS Patch II should be used when simulating hydrology and sediment yield for steep sloping watersheds (especially if average slope of the subwatershed is >25%) for more accurate simulation of hydrology and sediment using the SWAT model. The SWAT ArcView GIS Patch II is available at http://www.EnvSys.co.kr/~swat for free download.     

CHANGES IN HYDROLOGIC RESPONSE DUE TO STREAM NETWORK EXTENSION VIA LAND DRAINAGE ACTIVITIES1

ABSTRACT: The objective of this work is to determine the effects of extension of a stream network through land drainage activities during the late 1800s on the hydrologic response of a watershed. The Mackinaw River Basin in Central Illinois was chosen as the focus and the pre‐land and post‐land drainage activity hydrologic responses were obtained through convolution of the hill slope and channel responses and compared. The hill slope response was computed using the kinematic wave model and the channel response was determined using the geomorphologic instantaneous unit hydrograph method. Our hypothesis was that the hydrologic response of the basin would exhibit the characteristic effects of settlement (i.e., increases in peak discharges and decreases in times to peak). This, indeed, is what occurred; however, the increase in peak discharges diminishes as scale increases, leaving only the decrease in times to peak. At larger scales, the dispersive effects of the longer hill slope lengths in the pre‐settlement scenario seem to balance the depressive effects of the longer path lengths in the post‐settlement scenario, thus the pre‐settlement and post‐settlement peak discharges are approximately equivalent. At small scales, the dispersion caused by the hill slope is larger in the pre‐settlement case; thus, the post‐settlement peak discharges are greater than the pre‐settlement.     

Modeling Stream Channel Characteristics From Drainage‐Enforced DEMs in Puget Sound,Washington, USA1

Abstract: Mapping stream channels and their geomorphic attributes is an important step in many watershed research and management projects. Often insufficient field data exist to map hydromorphologic attributes across entire drainage basins, necessitating the application of hydrologic modeling tools to digital elevation models (DEMs) via a geographic information system (GIS). In this article, we demonstrate methods for deriving synthetic stream networks via GIS across large and diverse basins using drainage‐enforced DEMs, along with techniques for estimating channel widths and gradient on the reach scale. The two‐step drainage enforcement method we used produced synthetic stream networks that displayed a high degree of positional accuracy relative to the input streams. The accuracies of our estimated channel parameters were assessed with field data, and predictions of bankfull width, wetted width and gradient were strongly correlated with measured values (r² = 0.92, r² = 0.95, r² = 0.88, respectively). Classification accuracies of binned channel attributes were also high. Our methodology allows for the relatively rapid mapping of stream channels and associated morphological attributes across large geographic areas. Although initially developed to provide salmon recovery planners with important salmon habitat information, we suggest these methodologies are relevant to a variety of research and management questions.     

LARGE AREA HYDROLOGIC MODELING AND ASSESSMENT PART II: MODEL APPLICATION1

ABSTRACT: This paper describes the application of a river basin scale hydrologic model (described in Part I) to Richland and Chambers Creeks watershed (RC watershed) in upper Trinity River basin in Texas. The inputs to the model were accumulated from hydro-graphic and geographic databases and maps using a raster-based GIS. Available weather data from 12 weather stations in and around the watershed and stream flow data from two USGS stream gauge station for the period 1965 to 1984 were used in the flow calibration and validation. Sediment calibration was carried out for the period 1988 through 1994 using the 1994 sediment survey data from the Richland-Chambers lake. Sediment validation was conducted on a subwatershed (Mill Creek watershed) situated on Chambers Creek of the RC watershed. The model was evaluated by well established statistical and visual methods and was found to explain at least 84 percent and 65 percent of the variability in the observed stream flow data for the calibration and validation periods, respectively. In addition, the model predicted the accumulated sediment load within 2 percent and 9 percent from the observed data for the RC watershed and Mill Creek watershed, respectively.     

EFFECT OF WATERSHED SUBDIVISION ON SWAT FLOW,SEDIMENT, AND NUTRIENT PREDICTIONS1

ABSTRACT: The size, scale, and number of subwatersheds can affect a watershed modeling process and subsequent results. The objective of this study was to determine the appropriate level of subwatershed division for simulating flow, sediment, and nutrients over 30 years for four Iowa watersheds ranging in size from 2,000 to 18,000 km² with the Soil and Water Assessment Tool (SWAT) model. The results of the analysis indicated that variation in the total number of subwatersheds had very little effect on streamflow. However, the opposite result was found for sediment, nitrate, and inorganic P; the optimal threshold subwatershed sizes, relative to the total drainage area for each watershed, required to adequately predict these three indicators were found to be around 3, 2, and 5 percent, respectively. Decreasing the size of the subwatersheds below these threshold levels does not significantly affect the predicted levels of these environmental indicators. These threshold subwatershed sizes can be used to optimize input data preparation requirements for SWAT analyses of other watersheds, especially those within a similar size range. The fact that different thresholds emerged for the different indicators also indicates the need for SWAT users to assess which indicators should have the highest priority in their analyses.     

Impacts of land use conversion on bankfull discharge and mass wasting

Mass wasting and channel incision are widespread in the Nemadji River watershed of eastern Minnesota and northwestern Wisconsin. While much of this is a natural response to glacial rebound, sediment coring and tree ring data suggest that land use has also influenced these erosional processes. We characterized land use, inventoried mass wasting, surveyed stream channels and collected discharge data along segments of five streams in the Nemadji River watershed. Due to natural relief in this region, wetlands and agricultural lands are concentrated in the flatter terrain of the uplands of the Nemadji watershed, while forestland (coniferous or deciduous) is concentrated in the deeply incised (50-200% slope) stream valleys. Bankfull discharge was higher where forests had been converted from coniferous to deciduous forests and where there were fewer wetlands. Mass wasting increased exponentially with bankfull flows. While mass wasting was not correlated with forest type conversion and agricultural land use, it was negatively dependent upon wetland extent in headwater areas. Interactions between the spatial distribution of land use and terrain obfuscate any clear cause-and-effect relationships between land use, hydrology and fluvial processes.     

Using GIS to Delineate Headwater Stream Origins in the Appalachian Coalfields of Kentucky

Headwater streams have a significant nexus or physical, chemical, and/or biological connection to downstream reaches. Generally, defined as 1st‐3rd order with ephemeral, intermittent, or perennial flow regimes, these streams account for a substantial portion of the total stream network particularly in mountainous terrain. Due to their often remote locations, small size, and large numbers, conducting field inventories of headwater streams is challenging. A means of estimating headwater stream location and extent according to flow regime type using publicly available spatial data is needed to simplify this complex process. Using field‐collected headwater point of origin data from three control watersheds, streams were characterized according to a set of spatial parameters related to topography, geology, and soils. These parameters were (1) compared to field‐collected point of origin data listed in three nearby Jurisdictional Determinations, (2) used to develop a geographic information system (GIS)‐based stream network for identifying ephemeral, intermittent, and perennial streams, and (3) applied to a larger watershed and compared to values obtained using the high‐resolution National Hydrography Dataset (NHD). The parameters drainage area and local valley slope were the most reliable predictors of flow regime type. Results showed the high‐resolution NHD identified no ephemeral streams and 9 and 65% fewer intermittent and perennial streams, respectively, than the GIS model.     

Computer programs for calculating soil loss on a watershed basis

A series of computer programs designed to predict gross annual soil loss on a watershed basis by application of the Universal Soil Loss Equation (USLE) have been developed. The programs provide an easy-to-use, flexible, and standardized means of organizing base data and applying the USLE to large land areas. The programs can be used to assess and to evaluate the effects of changing land-use patterns and conservation practices on soil losses. Critical or problem areas can be readily identified. The USLE Computer Programs are a useful research tool for investigators involved in water quality management, 208 planning, or conservation research.The package of computer programs consists of three main components: data input, the Main Program, and the Totals Program. Input data include both field base data describing the watershed and corresponding values for the factors in the USLE. The Main Program calculates the average rate of soil loss (tons/ acre/yr) and the total soil loss (tons/yr) for the smallest subunit of the watershed identified as the soil unit. Also calculated is an RKLS factor, which is an indication of the erosive potential of a given soil type, slope, and slope length, under a particular rainfall regime. The Totals Program aggregates soil unit losses into progressively larger units, that is, field, farm, subwatershed, and watershed units. An example of the programs' versatility and use is presented.     

WATER QUALITY MODEL OUTPUT UNCERTAINTY AS AFFECTED BY SPATIAL RESOLUTION OF INPUT DATA1

ABSTRACT: Resolution of the input GIS data used to parameterize distributed‐parameter hydrologic/water quality models may affect uncertainty in model outputs and impact the subsequent application of model results in watershed management. In this study we evaluated the impact of varying spatial resolutions of DEM, land use, and soil data (30 × 30 m, 100 × 100 m, 150 × 150 m, 200 × 200 m, 300 × 300 m, 500 × 500 m, and 1,000 × 1,000 m) on the uncertainty of SWAT predicted flow, sediment, NO₃‐N, and TP transport. Inputs included measured hydrologic, meteorological, and watershed characteristics as well as water quality data from the Moores Creek watershed in Washington County, Arkansas. The SWAT model output was most affected by input DEM data resolution. A coarser DEM data resolution resulted in decreased representation of watershed area and slope and increased slope length. Distribution of pasture, forest, and urban areas within the watershed was significantly affected at coarser resolution of land use and resulted in significant uncertainty in predicted sediment, NO₃‐N, and TP output. Soils data resolution had no significant effect on flow and NO₃‐N predictions; however, sediment was overpredicted by 26 percent, and TP was underpredicted by 26 percent at 1,000 m resolution. This may be due to change in relative distribution of various hydrologic soils groups (HSGs) in the watershed. Minimum resolution for input GIS data to achieve less than 10 percent model output error depended upon the output variable of interest. For flow, sediment, NO₃‐N, and TP predictions, minimum DEM data resolution should range from 30 to 300 m, whereas minimum land use and soils data resolution should range from 300 to 500 m.     

EVALUATION OF THE ROSGEN STREAM CLASSIFICATION SYSTEM IN CHEQUAMEGON‐NICOLET NATIONAL FOREST,WISCONSIN1

ABSTRACT: Data collected from 121 stream reaches during 1991 to 1993 were evaluated to determine the applicability of the Rosgen Stream Classification System (RSCS) to the low relief terrain within the Chequamegon‐Nicolet National Forest (CNNF) in Wisconsin, USA. All reaches were classified to RSCS Level I and II except that 10.7 percent had sinuosities below the continuum limits and one reach had a predominantly organic substrate. Five of eight possible RSCS Level I types were observed including B, C, D, A, E, and F; 86 percent were C and E types. Seventeen of 94 possible RSCS Level II types were observed. Most reaches were slightly entrenched, had low to moderate width/depth ratios, relatively low sinuosity, low slope, and sand or gravel as the dominant channel material. Discriminant analyses were used to verify the applicability of RSCS for streams within the CNNF; discriminant functions correctly classified 92.5 and 94.7 percent of the Level I and II RSCS types, respectively. When limits for E and F types were modified slightly at Level II by adding an additional category for slopes less than 0.1 percent (a modification we recommend for low relief terrain), discriminant functions correctly classified 99.1 percent of the types. Adding another slope break at 0.3 percent produced similar results. Based on our analyses, RSCS works well within the CNNF and is probably applicable to other areas with low‐relief terrain.     

ESTIMATING DRAINAGE AREA ON STEEP UNIFORM SLOPES1

ABSTRACT: A study was conducted to derive a simple procedure for estimating the tributary drainage area that may potentially develop at the base of a uniformly graded, steep slope. Data were extracted from a nil development flume study in which 112 rills and their tributary drainage areas were documented. The tributary drainage area for each rill was correlated to the slope length and the slope gradient. The findings are considered applicable to planar, convex, and divergent slopes with lengths of up to 130 m. Field verification is needed to extend the applicability of the relation.     

NONPOINT SOURCE POLLUTION POTENTIAL IN AN AGRICULTURAL WATERSHED IN NORTHWESTERN PENNSYLVANIA1

ABSTRACT: A 155,947 ha portion of the Shenango River watershed in western Pennsylvania was evaluated as to the potential impact of agriculture drainage on water quality. Approximately a third of the area is being used as either cropland or pasture with approximately an equal percentage in forest lands. Eleven subwatersheds were evaluated as to their potential for nonpoint source pollution according to the criteria established by the Pennsylvania Department of Environmental Resources for the Chesapeake Bay Pollution Abatement Program. The individual components and overall rating for each subwatershed were then evaluated as to their correlation with four water quality variables based on 104 samples collected at 26 sampling stations throughout the watershed. There was a significant correlation between the overall rating factor for each subwatershed and each of the four water quality variables. In general, the watershed delivery factor, animal nutrient factor, and management factors were correlated with fecal coliform and phosphorus in the receiving streams, whereas the ground water delivery factor appeared to be more important in determining nitrate concentrations in these streams. These results indicate that manure and nutrient management, along with the exclusion of livestock from streams and the enhancement and/or replacement of riparian wetlands, are important approaches in reducing agricultural impacts in fresh water ecosystems.     

SENSITIVITY CONSIDERATIONS WHEN MODELING HYDROLOGIC PROCESSES WITH DIGITAL ELEVATION MODEL1

ABSTRACT: The purpose of this paper is to investigate the sensitivity of a hydrologic models to the type of DEM used. This was done while modeling basin water quality with 1:24,000 and 1:250,000 U.S. Geological Survey DEMs as input to model hydro‐logical processes. The manner in which the model results were sensitive to the choice of raster cell size (scale) is investigated in this study. The Broadhead watershed, located in New Jersey, USA, was chosen as a study area. Curve numbers were estimated by a trial and error to match simulated and observed total discharge. Monthly runoff for the watershed was used in the calibration process. Higher runoff volumes were simulated by the model when the 1:24,000 DEM were used as input data, probably due to the finer resolution which simulated increased average slope and hence higher estimated runoff from the watershed. As the simulated slope of the watershed is flatten with the 1:250,000 DEM, the response of stream flow was delayed and simulated less runoff volume.     

VARIATION OF DRAINAGE DENSITY IN A SMALL BRITISH COLUMBIA WATERSHED1

ABSTRACT: Short-term fluctuations in stream length were monitored in a small watershed located 10 kilometers east of Vancouver, British Columbia. Variation in discharge was reflected in a log-normal relationship with expansion and contraction of the stream net. The analysis of the variation of drainage density data using the individual hydrograph revealed a hysteresis effect. The stream net (as measured by drainage density) expanded more rapidly on the rising limb than it contracted on the recessional limb.