30-m Digital Elevation Model Research Articles

The Apache Longbow–Hellfire Missile Test at Yuma Proving Ground: Ecological Risk Assessment for Tracked Vehicle Movement across Desert Pavement

ABSTRACT A multiple stressor risk assessment was conducted at Yuma Proving Ground, Arizona, as a demonstration of the Military Ecological Risk Assessment Framework (MERAF). The focus was a testing program at Cibola Range that involved an Apache Longbow helicopter firing Hellfire missiles at moving targets, that is, M60-A1 tanks. This article describes the ecological risk assessment, using the MERAF, for the tracked vehicle movement component of the testing program. The principal stressor associated with tracked vehicle movement was soil disturbance, and a resulting, secondary stressor was hydrological change. Water loss to desert wash vegetation was hypothesized to result from increased infiltration and/or evaporation associated with vehicle disturbances to surrounding desert pavement, potentially affecting mule deer as well as vegetation. The simulated exposure of wash vegetation to water loss was quantified using estimates of disturbed land area from a digital orthogonal quarter quadrangle aerial photo and field observations, a 30-m digital elevation model, the flow accumulation feature of ESRI ArcInfo GIS, and a two-step runoff process dependent on soil characteristics and the extent of disturbance. In all simulated scenarios, the absolute amount of water lost increased with distance from the disturbance downslope in the washes; however, the percentage of water lost was greatest in land areas immediately downslope of a disturbance. Potential effects on growth and survival of desert wash trees were quantified by comparing water availability from the hydrologic model to water volume thresholds required for wash trees to survive and persist, derived from a local study. For both the incremental risk of the test program and for the combination of test and pretest disturbances, this demonstration of MERAF found no significant risk to either wash vegetation growth and survival or mule deer abundance and reproduction.

Desert terrain characterization of landforms and surface materials within vehicle test courses at U.S. Army Yuma Proving Ground, USA

The U.S. Army Yuma Proving Ground is the Department of Defense desert environment test center within the Sonoran Desert of Arizona. The Yuma Proving Ground has ∼320 km of unpaved vehicle test courses that cross a variety of landforms of diverse geologic age and characteristics. The surface materials of the courses ranges from bedrock to silt and their topography varies from steep and rolling to flat. Research presented here aims to provide a systematic characterization of the terrain of eight vehicle endurance and three dust courses so that their comparability with other desert areas of the World may be assessed. Landform and surface cover (upper 1 m) characterization was accomplished by geomorphic mapping based on 1-m resolution IKONOS satellite imagery, 10-m digital elevation models, field verification, and by assimilating pre-existing soil surveys and geologic maps, as well as site-specific investigations. Results provide an assessment of each test course, including information on the landform, geology, surface materials, soil type, degree of desert pavement development, dust content, and percent slope. Data is presented both on individual terrain property maps for each course and in the form of tabulated data for each official milepost marker along the courses. The results for one course area and an example of how they may be used to assess comparability with another desert of interest are presented here with the objective to improve the fidelity of desert testing during material research, development, testing and evaluation prior to deployment in the field.

Soil erosion prediction using the Revised Universal Soil Loss Equation (RUSLE) in a GIS framework, Chania, Northwestern Crete, Greece

Soil erosion is a growing problem in southern Greece and particularly in the island of Crete, the biggest Greek island with great agricultural activity. Soil erosion not only decreases agricultural productivity, but also reduces the water availability. In the current study, an effort to predict potential annual soil loss has been conducted. For the prediction, the Revised Universal Soil Loss Equation (RUSLE) has been adopted in a Geographical Information System framework. The RUSLE factors were calculated (in the form of raster layers) for the nine major watersheds which cover the northern part of the Chania Prefecture. The R-factor was calculated from monthly and annual precipitation data. The K-factor was estimated using soil maps available from the Soil Geographical Data Base of Europe at a scale of 1:1,000,000. The LS-factor was calculated from a 30-m digital elevation model. The C-factor was calculated using Remote Sensing techniques. The P-factor in absence of data was set to 1. The results show that an extended part of the area is undergoing severe erosion. The mean annual soil loss is predicted up to ∼200 (t/ha year−1) for some watersheds showing extended erosion and demanding the attention of local administrators.

Application of Remote Sensing and GIS for Improving Rice Production in Flood-prone Areas: A Case Study in Lower Chi-River Basin, Thailand

Rice is annually destroyed by seasonal flooding in South and Southeast Asia. In order to reduce the loss of rice production, an accurate assessment of flood-affected paddies is essential. Taking the 2001 monsoon flooding that hit the Lower Chi River Basin, northeastern Thailand, as a case study, we derived the flood depth over the large and inaccessible areas using a RADARSAT- 1 image acquired during the flooding peak and a 30-m Digital Elevation Model. Based on the elongation properties of rice, we used the critical water-depth of 80 cm to classify the flood-affected paddies into 'damaged' or 'non-damaged' paddies. Our results show that only about 40% of the flood-affected paddies were actually damaged paddies. In addition, we analyzed the damaged paddies by overlaying the derived water depth with the flood frequency map in order to recommend planting the appropriate rice varieties. Our analysis reveals that remote sensing and GIS can be used to minimize the loss of rice production and thus help farmers to increase their income in the long run, which will save the government money paid in flood compensations.

북향개방지수가 복잡지형의 일 최고기온 추정오차 저감에 미치는 영향

The normalized difference in incident solar energy between a target surface and a level surface (overheating index, OHI) is useful in eliminating estimation error of site-specific maximum temperature in complex terrain. Due to the complexity in its calculation, however, an empirical proxy variable called northern exposure index (NEI) which combines slope and aspect has been used to estimate OHI based on empirical relationships between the two. An experiment with real-world landscape and temperature data was carried out to evaluate performance of the NEI - derived OHI (N-OHI) in reduction of spatial interpolation error for daily maximum temperature compared with that by the original OHI. We collected daily maximum temperature data from 7 sites in a mountainous watershed with a area and a 795m elevation range () in Pyongchang, Kangwon province. Northern exposure index was calculated for the entire 166,050 grid cells constituting the watershed based on a 30-m digital elevation model. Daily OHI was calculated for the same watershed ana regressed to the variation of NEI. The regression equations were used to estimate N-OHI for 15th of each month. Deviations in daily maximum temperature at 7 sites from those measured at the nearby synoptic station were calculated from June 2006 to February 2007 and regressed to the N-OHI. The same procedure was repeated with the original OHI values. The ratio sum of square errors contributable by the N-OHI were 0.46 (winter), 0.24 (fall), and 0.01 (summer), while those by the original OHI were 0.52, 0.37 and 0.15, respectively.

Relating crop yield to topographic attributes using Spatial Analysis Neural Networks and regression

Land-surface topographic attributes can be useful for estimating stable spatial patterns of crop yield caused by spatial variability in soils and water availability. We present spatial analyses of grain yield for three fields of dryland winter wheat in northeastern Colorado using topographic attributes as explanatory variables. Topographic attributes including elevation, slope, aspect, curvature, specific contributing area, and wetness index were computed from a 10-m digital elevation model. A Spatial Analysis Neural Network (SANN) algorithm was used for joint spatial interpolation and yield prediction from the topographic attributes. SANN prediction errors were compared with the results of Multiple Linear Regression (MLR). SANN and MLR were assessed in terms of bias and relative root mean squared error (rRMSE) using validation data. SANN out-performed MLR in multivariate estimation, but not for the univariate cases. The greatest advantage of SANN was seen using four or more topographic attributes, whereas MLR showed diminishing efficiency with more than three explanatory variables. Prediction/interpolation errors within a given field were reduced substantially by using the spatial coordinates (latitude and longitude) in tandem with topographic attributes. The rRMSE value reached a minimum of 0.44 (model efficiency, E = 0.80) for interpolation with SANN on the West field. Using only topographic attributes as input, the minimum rRMSE values were 0.59 ( E = 0.65) for SANN with 5 variables and 0.72 ( E = 0.48) for MLR with 4 or 5 explanatory variables. Thus, this study demonstrated the utility of SANN with topographic attributes that contain implicit soil and water information for estimating spatial patterns of dryland crop yield.

Incorporating Soil Surface Erosion Prediction into Forest Road Alignment Optimization

A previous study introduced a forest road design model developed to simultaneously optimize horizontal and vertical alignments of forest roads using a Tabu Search optimization technique and a high-resolution Digital Elevation Model (DEM). In this study, surface erosion prediction was incorporated into the road design model, so that users can optimize horizontal and vertical alignments of forest roads while constrained by maximum allowable sediment delivery from roads to streams. The road alignment optimization model was applied to a part of the Capitol State Forest in western Washington state. The application confirms the potential of the model to determine forest road alignments in a way to reduce total road costs as well as sediment delivery to streams. This paper also discusses the effects of DEM resolution on forest road alignment optimization. The accuracy of generating ground profile and forest road alignments depends on the resolution and accuracy of the DEM. The study results suggest that a 10-m grid DEM might be inappropriate to use for the purpose of road design and alignment optimization due to the lower accuracy in its elevation representation.

Verification of the Vertical Error in C-Band SRTM DEM Using ICESat and Landsat-7, Otter Tail County, MN

The Shuttle Radar Topography Mission (SRTM) provided scientists with digital elevation data on a nearly global scale and with highly consistent accuracy. This paper compares elevation values of the C-band SRTM 30-m digital elevation model (DEM) with pointwise elevations from the Ice, Cloud, and land-Elevation Satellite (ICESat) laser altimetry for Otter Tail County, Minnesota. The accuracy of SRTM DEM is measured as a function of land covers and geomorphologic characteristics. The typical mean vertical difference between the SRTM DEM and ICESat elevations in this paper was determined in each classified land-use type and is approximately 1.5 m over bare ground, with the SRTM measuring lower elevations. Significant changes in the SRTM DEM uncertainties have been identified over different surface types classified from Landsat-7 imagery, e.g., bare ground, urban, and forested areas. Based on this result, the difference of the SRTM 30-m DEM and ICESat elevations has been removed from the DEM and made available for improved hydrological applications

TOPOGRAPHIC INDICES

Real-time kinematic (RTK) global positioning system (GPS) receivers are being marketed for applications where accurate estimates of elevation are required. Because this technology is fairly expensive, the quality of these data should be considered with respect to that of USGS digital elevation DEMs (digital elevation models) that are freely available for most of the United States at the 30-m grid scale and for some areas on 10-m grids. The objective of this study was to compare terrain attributes calculated from USGS DEMs with those calculated from DEMs developed with precision RTK GPS measurements. At two locations in central Kentucky, intensive elevation measurements were collected with RTK GPS. The USGS and RTK data for these areas were interpolated to create 4-m DEMs for comparing terrain attributes. Correlations between the attributes developed from different DEMs were substantially larger for slope and aspect than for curvature, topographic wetness index, and stream power index values. Although there were large quantitative differences in terrain attribute values, the general spatial patterns were similar for the attributes calculated from the RTK and USGS DEMs. However, landscape features defined using DEMs developed with RTK data were more similar to those observed with 10- than 30-m USGS DEMs. For some course-scale applications, 30-m USGS DEMs may be adequate for identifying important landscape characteristics. However, the resolution of many of these features will be substantially better with 10-m USGS DEMs. However, quantitative estimates of terrain attributes at a specific location may be grossly incorrect. For situations where accuracy is essential, precision elevation surveys will be required.

Two models for evaluating landslide hazards

Two alternative procedures for estimating landslide hazards were evaluated using data on topographic digital elevation models (DEMs) and bedrock lithologies in an area adjacent to the Missouri River in Atchison County, Kansas, USA. The two procedures are based on the likelihood ratio model but utilize different assumptions. The empirical likelihood ratio model is based on non-parametric empirical univariate frequency distribution functions under an assumption of conditional independence while the multivariate logistic discriminant model assumes that likelihood ratios can be expressed in terms of logistic functions. The relative hazards of occurrence of landslides were estimated by an empirical likelihood ratio model and by multivariate logistic discriminant analysis. Predictor variables consisted of grids containing topographic elevations, slope angles, and slope aspects calculated from a 30-m DEM. An integer grid of coded bedrock lithologies taken from digitized geologic maps was also used as a predictor variable. Both statistical models yield relative estimates in the form of the proportion of total map area predicted to already contain or to be the site of future landslides. The stabilities of estimates were checked by cross-validation of results from random subsamples, using each of the two procedures. Cell-by-cell comparisons of hazard maps made by the two models show that the two sets of estimates are virtually identical. This suggests that the empirical likelihood ratio and the logistic discriminant analysis models are robust with respect to the conditional independent assumption and the logistic function assumption, respectively, and that either model can be used successfully to evaluate landslide hazards.

Effects of Digital Elevation Model Errors on Spatially Distributed Seismic Slope Stability Calculations: An Example from Seattle, Washington

More than 1,600 centimeter-accurate differential GPS measurements show that elevation errors in a 10-m digital elevation model (DEM) covering an 18-km 2 portion of Seattle can have significant effects on calculated values of slope angle, static factor of safety, Newmark acceleration, and slope displacement. Declustered error data from the study area have a mean of ! 0.92 m and a standard deviation of 62.36 m, comparable to results reported by others. Conditional Gaussian simulation and normal score back-transformation were used to create 25 error realizations with the same statistical distribution and spatial correlation as the measured errors, and the error realizations were subsequently used to create elevation, slope angle, static factor of safety, Newmark acceleration, and displacement realizations. The maximum slope angle uncertainty magnitude decreases frommorethan108toabout48asslopeangleincreases, but the minimum slope angle uncertainty varies between 618 and 628 regardless of slope angle. Typical slope angle standard deviations arising from DEM errors are in the range of 638 to 648. The resulting uncertainties in calculated Newmark displacements as the consequence of a hypothetical large earthquake affecting the study area (Arias intensity of 4 m/s) are in most cases less than 63 cm, but some values are greater than 610 cm. If the acceptable maximum probability of misclassifying a potentially unstable area as stable is decreased from 50 to 5 percent, incorporation of displacement uncertainty into the calculusofhazardincreasesfrom0.6to4.1percentthe portion of the study area that may be susceptible to landsliding as a result of the hypothetical earthquake.

Spatial structures of stream and hillslope drainage networks following gully erosion after wildfire

AbstractThe drainage networks of catchment areas burned by wildfire were analysed at several scales. The smallest scale (1–1000 m2) representative of hillslopes, and the small scale (1000 m2 to 1 km2), representative of small catchments, were characterized by the analysis of field measurements. The large scale (1–1000 km2), representative of perennial stream networks, was derived from a 30‐m digital elevation model and analysed by computer analysis.Scaling laws used to describe large‐scale drainage networks could be extrapolated to the small scale but could not describe the smallest scale of drainage structures observed in the hillslope region. The hillslope drainage network appears to have a second‐order effect that reduces the number of order 1 and order 2 streams predicted by the large‐scale channel structure. This network comprises two spatial patterns of rills with width‐to‐depth ratios typically less than 10. One pattern is parallel rills draining nearly planar hillslope surfaces, and the other pattern is three to six converging rills draining the critical source area uphill from an order 1 channel head. The magnitude of this critical area depends on infiltration, hillslope roughness and critical shear stress for erosion of sediment, all of which can be substantially altered by wildfire. Order 1 and 2 streams were found to constitute the interface region, which is altered by a disturbance, like wildfire, from subtle unchannelized drainages in unburned catchments to incised drainages. These drainages are characterized by gullies also with width‐to‐depth ratios typically less than 10 in burned catchments. The regions (hillslope, interface and channel) had different drainage network structures to collect and transfer water and sediment. Copyright © 2005 John Wiley & Sons, Ltd.

Horizontal resolution and data density effects on remotely sensed LIDAR-based DEM

Terrain analysis of digital elevation models (DEM) has become an important technique to assess landscape and watershed scale hydrologic and pedologic processes and the spatial variability of soil and ecologic properties. Light detecting and ranging (LIDAR) elevation data sets provide the flexibility needed to produce multiple horizontal resolutions of DEM from the same data source. A series of 61 LIDAR tiles (100 ha) were collected from the North Carolina Flood Mapping Program covering the spatial extent of the Hofmann Forest in the Lower Coastal Plain of Eastern North Carolina. The LIDAR data set was reduced to 50%, 25%, 10%, 5%, and 1% of the original density. We created 5-, 10-, and 30-m DEM with 0.1 m vertical precision for each density level and used paired t-test to determine if the true mean of their differences were equal to zero. Differences indicated that for the 30-m DEM, LIDAR data sets could be reduced to 10% of their original data density without statistically altering the produced DEM. However, the 10-m DEM could only be reduced to 25% of the original data set before statistically altering the DEM. Data reduction was more limited for the 5-m DEM with possible reduction only to 50% of their original density without producing statistically different DEM. Our evaluation provides some indication as to the minimum required LIDAR data density to produce a DEM of a given horizontal resolution. However, evaluation of additional horizontal resolutions and additional density reduction is required to provide a clearer understanding of the effect of LIDAR data density.

Lidar Elevation Data for Surface Hydrologic Modeling: Resolution and Representation Issues

This paper is concerned with the application of high spatial resolution elevation data derived from light detection and ranging technologies (lidar) to surface hydrologic modeling. In recent years, airborne lidar technology has been employed to develop high accuracy digital elevation models (DEMs) with horizontal resolution on the order of a few meters. As with any spatial data product there are limits to the lidar's practical use that vary with the intended application. This paper considers potential issues and challenges for the use of lidar-derived DEMs in surface hydrologic modeling applications, such as characterizing flow direction and power, identifying sub-basins in a watershed, and calculating upstream contributing area and other variables. We compare results using conventional 30m DEMs and 6m lidar for a high relief study area and a low relief study area. Results are more comparable between these data sources, regardless of hydrologic operation, for the high relief area, while the similarity of results in the low relief area is dependent upon the particular operation. Post-processing on the lidar data successfully removed such flow obstacles as bridges that might have artificially impeded surface flow. An exploration of the effect of spatial resolution on results suggests that cell size is a more significant factor than production method.

The occurrence of alpine permafrost in the Front Range of Colorado

Permafrost distribution, or ground that remains frozen for at least 2 years, has been modeled using a combination of Geographic Information System (GIS) techniques, Digital Elevation Model (DEM) variables, and land cover in alpine regions of the world. In the Front Range, however, no such empirical models have been developed, and field data are restricted in spatial extent, but rock glaciers are in abundance. Here, I present a probabilistic logistic regression model that is based on topoclimatic information (elevation and aspect) for rock glaciers derived from U.S. Geological Survey (USGS) 10-m DEMs. Classes of land cover, obtained from an Enhanced Thematic Mapper Plus (ETM+) image classification, were assigned weights and were then multiplied by the regression results to refine estimates. The effectiveness of the model was evaluated by comparing mean probability scores with rock glacier activity categories, Mean Annual Air Temperature (MAAT) from climatic stations on Niwot Ridge, and Bottom Temperature of winter Snow (BTS) measurements, while a Monte Carlo simulation was used to detect uncertainty associated with the original DEM. Permafrost scores >50% covered about 8.9% (242 km 2) of the study area (2722 km 2) with the highest scores clustered around Longs and Rowe Peaks. Permafrost locations showed a strong correlation with rock glacier activity classes, the −1.0 °C MAAT isotherm, and BTS measurements less than −3.0 °C. The uncertainty analysis revealed that slight global differences exist between the original and error prone DEM; however, local variations in aspect caused the most uncertainty. These results indicate that the model accurately represents regional distribution of permafrost. Therefore, topoclimatic information from rock glaciers and land cover, when combined with an uncertainty analysis, can effectively be used to map the occurrence of Front Range permafrost, providing an imperative tool for cartographers, planners, and geocryologists.

Identification of Mapped Ice-Margin Positions in Western New York from Digital Terrain-Analysis and Soil Databases

Geographic Information Systems (GIS) are rarely applied to problems associated with Wisconsinan ice-margins. This study identifies inconsistencies in ice extent in the Salamanca Re-entrant in western New York as mapped using soil properties and surficial geology. In essence, previous studies have revealed a zone of disagreement between those areas that were clearly glaciated and those that were not. This study uses a raster-based approach to extract the soil pH, silt, and clay contents from digital soil databases; and the morphometric parameters of elevation, slope, terrain ruggedness, and planform and profile curvature from mosaiced 10-m digital elevation models. Two-way ANOVA indicates a close correspondence between the zone of disagreement and the glaciated area when soil information is used; however on the basis of morphometry, the area of disagreement is analogous to the unglaciated terrain. These results highlight both previous difficulties and the source of the mapped differences, suggesting that a GIS analysis of former ice margins is a productive preliminary step to their precise delineation.

Landscape Modeling of In Situ Soil Respiration in a Forested Watershed of Southeastern Kentucky, USA

Data on carbon fluxes such as soil respiration are needed to develop strategies for increased carbon sequestration and reduced levels of atmospheric trace gases. Soil respiration is driven by proximal factors (e.g., soil temperature and soil moisture), which affect soil respiration by regulating microbial and root activity, and distal factors (e.g., topography), which affect soil respiration by influencing proximal factors. The objectives of this study were: (1) to relate measured carbon dioxide flux from forest soils to proximal and distal factors; and (2) to develop predictive soil-landscape models of soil respiration across a forested watershed. Carbon dioxide flux from the soil surface was measured monthly for 12 months at sampling points selected using a random stratified approach, with strata established based on slope aspect (NE and SW), slope shape (concave and convex), and slope position (upper, middle, and lower backslope). We generated empirical models using robust linear regression techniques to examine relationships between soil respiration and both proximal (soil physical and chemical properties) and distal factors (terrain attributes calculated from a 30-m digital elevation model). In situ soil respiration rates were greater on the NE-facing slopes than the SW-facing slopes. The models that we developed and validated explain up to 66% of variability in measured soil respiration, although seasonal model relationships varied. Soil temperature was the most consistent proximal factor for predicting soil respiration rates, while slope aspect was the most consistent distal factor among the models.

Wildfire-related debris-flow initiation processes, Storm King Mountain, Colorado

A torrential rainstorm on September 1, 1994 at the recently burned hillslopes of Storm King Mountain, CO, resulted in the generation of debris flows from every burned drainage basin. Maps (1:5000 scale) of bedrock and surficial materials and of the debris-flow paths, coupled with a 10-m Digital Elevation Model (DEM) of topography, are used to evaluate the processes that generated fire-related debris flows in this setting. These evaluations form the basis for a descriptive model for fire-related debris-flow initiation. The prominent paths left by the debris flows originated in 0- and 1st-order hollows or channels. Discrete soil-slip scars do not occur at the heads of these paths. Although 58 soil-slip scars were mapped on hillslopes in the burned basins, material derived from these soil slips accounted for only about 7% of the total volume of material deposited at canyon mouths. This fact, combined with observations of significant erosion of hillslope materials, suggests that a runoff-dominated process of progressive sediment entrainment by surface runoff, rather than infiltration-triggered failure of discrete soil slips, was the primary mechanism of debris-flow initiation. A paucity of channel incision, along with observations of extensive hillslope erosion, indicates that a significant proportion of material in the debris flows was derived from the hillslopes, with a smaller contribution from the channels. Because of the importance of runoff-dominated rather than infiltration-dominated processes in the generation of these fire-related debris flows, the runoff-contributing area that extends upslope from the point of debris-flow initiation to the drainage divide, and its gradient, becomes a critical constraint in debris-flow initiation. Slope-area thresholds for fire-related debris-flow initiation from Storm King Mountain are defined by functions of the form A cr(tan θ) 3= S, where A cr is the critical area extending upslope from the initiation location to the drainage divide, and tan θ is its gradient. The thresholds vary with different materials.

Digital elevation model resolution: effects on terrain attribute calculation and quantitative soil-landscape modeling

The accuracy of digital elevation models (DEM) and DEM-derived products depends on several factors, including the horizontal resolution and vertical precision at which the elevation data are represented, and the source of the elevation data. This accuracy becomes increasingly important as we extend the use of DEM data for spatial prediction of soil attributes. Our objective was to compare terrain attributes and quantitative soil-landscape models derived from grid-based DEM represented at different horizontal resolutions (10 and 30 m), represented at different vertical precisions (0.1 and 1 m), and acquired from different sources. Decreasing the horizontal resolution of the field survey DEM produced lower slope gradients on steeper slopes, steeper slope gradients on flatter slopes, narrower ranges in curvatures, larger specific catchment areas in upper landscape positions, and lower specific catchment areas values in lower landscape positions. Overall, certain landscape features were less discernible on the 30-m DEM than on the 10-m DEM. Decreased vertical precision produced a large proportion of points with zero slope gradient and zero slope curvature, and a large number of steeply sloping and more highly curved areas. Differences among DEM from different sources were more significant, with less accurate representation of depressions and drainage pathways with the USGS DEM as compared to the field survey DEM. Empirical models developed from different DEM included similar predictive terrain attributes, and were equally successful in predicting A-horizon depth (AHD) in the validation data set.

Assessing landslide potential using GIS, soil wetness modeling and topographic attributes, Payette River, Idaho

This study utilizes GIS modeling to determine if the location of 559 landslides in the 875 km 2 catchment of the Middle Fork of the Payette River, Idaho can be predicted based on topographic attributes and a wetness index generated by the DYNWET model. Slope and elevation were significantly related to landslide occurrence at this landscape scale. Aspect was also retained as a variable for further analysis because, despite a non-significant chi-square relation to landslide occurrence, graphical analysis suggested a relation between aspect and mass wasting. Chi-square analysis indicated that plan and profile curvature, flow path length, upslope contributing area, and the DYNWET-based moisture index were not significantly related to landsliding. A Bayesian probability model based on combinations of elevation, slope, aspect, and wetness indicates that elevation exhibits the closest relation to landsliding, followed by slope; but that neither aspect nor wetness index values help in prediction. The Bayesian probability model using elevation and slope generates a map of relative landslide risk that can be used to direct activities away from mass wasting prone areas. The association between elevation and landslides is perplexing but is perhaps due to the location of logging road at specific elevations (roads could not be included in the input data for analysis because they have not been adequately mapped). The lack of explanation provided by the DYNWET wetness index was also surprising and may be due to the 30-m digital elevation model (DEM) and the soils data having resolutions too coarse to adequately portray local variations key to mass wasting. We believe the inadequacy of data to drive the models is typical of the majority of catchment scale setting. For now, the ability of researchers to effectively model landscape scale landsliding is more limited by the type, resolution, and quality of available data than by the quality of the landslide models.