Guangxing Wang

Improvement in mapping vegetation cover factor for the universal soil loss equation by geostatistical methods with Landsat Thematic Mapper images

The universal soil loss equation (USLE) is a product of six factors: (1) rainfall erosivity, (2) soil erodibility, (3) slope length, (4) slope steepness, (5) cover and management, and (6) support practice, and is widely used to estimate average annual soil loss. The cover and management variable, called the C factor, represents the effect of cropping and management practices on erosion rates in agriculture, and the effect of ground, tree and grass canopy covers on reduction of soil loss in non-agriculture situation. This study compared three traditional and three geostatistical methods for mapping the C factor. They included vegetation classification with average, linear and log-linear regression for C factor assignment, sequential Gaussian cosimulations with and without Thematic Mapper (TM) images, and colocated cokriging with TM images. The coefficient of correlation between estimates and observations varied from 0.4888 to 0.7317, and the root mean square error (RMSE) from 0.0159 to 0.0203. The sequential Gaussian cosimulation with a TM ratio image resulted in the highest correlation and the smallest RMSE, and reproduced the best and most detailed spatial variability of the C factor. This method may thus be recommended for mapping the C factor. It is also expected that this method could be applied to image-based mapping in other disciplines.

Mapping Multiple Variables for Predicting Soil Loss by Geostatistical Methods with TM Images and a Slope Map

Soil erosion is widely predicted as a function of six input factors, including rainfall erosivity, soil eredibility, slope length, slope steepness, cover management, and support practice. Because of the multiple factors, their interactions, and their spatial and temporal variability, accurately mapping the factors and further soil loss is very difficult. This paper compares two geostatistical methods and a traditional stratification to map the factors and to estimate soil loss. Soil loss is estimated by integrating a sample ground data set, TM images, and a slope map. The geostatistical methods include collocated cokriging and a joint sequential co-simulation model. With both geostatistical methods, local estimates and variances at any location cohere the factors and soil loss are unknown can be computed. The results showed that the two geostatistical methods performed significantly better than traditional stratification in terms of overall and spatially explicit estimate. Furthermore, the cokriging led to higher accuracy of mean estimates than did the co-simulation, while the latter provided decision makers with reliable uncertainties of the local estimates as useful information to assess risk when making decisions based on the prediction maps.

Spatial and temporal prediction and uncertainty of soil loss using the revised universal soil loss equation: a case study of the rainfall-runoff erosivity R factor

Abstract Soil loss is commonly predicted using the revised universal soil loss equation consisting of rainfall–runoff erosivity, soil erodibility, slope steepness and length, cover management, and support practice factors. Because of the multiple factors, their interactions, and spatial and temporal variability, soil erosion varies considerably over space and time. For these reasons, modeling soil loss is very complicated. Decision-makers need local and regional estimates of soil loss as well as their corresponding uncertainties. Neglecting the local and detailed information may lead to improper decision-making. This paper demonstrates a strategy based on a sample data set and a geostatistical method called sequential Gaussian simulation to derive local estimates and their uncertainties for the input factors of a soil erosion system. This strategy models the spatial and temporal variability of the factors and derives their estimates and variances at any unknown location and time. This strategy was applied to a case study at which the rainfall–runoff erosivity R factor was spatially and temporally estimated using a data set of rainfall. The results showed that the correlation between the observations and estimates by the strategy ranged from 0.89 to 0.97, and most of the mean estimates fell into their confidence intervals at a probability of 95%. Comparing the estimates of the R factor using a traditional isoerodent map to the observed values suggested that the R factor might have increased and a new map may be needed. The method developed in this study may also be useful for modeling other complex ecological systems.

Uncertainty assessment of soil erodibility factor for revised universal soil loss equation

Soil erodibility accounts for the influence of the intrinsic soil properties on soil erosion and is one of six factors in the Revised Universal Soil Loss Equation (RUSLE), a most widely used model to predict long-term average annual soil loss. In a traditional soil survey, each of the soil types (classes) is assigned with a soil erodibility value that is assumed to be constant over time. However, heterogeneity of soil in time and in space tends to support the concept that soil erodibility depends dynamically and spatially on the set of properties of a specific soil. This study statistically compared the published soil erodibility values with those from a set of soil samples in terms of their differences. The published values tend to underestimate soil erodibility. This feature is also supported by the uncertainty assessment in difference maps of the published K values versus those from soil samples. Spatial prediction and uncertainty analysis of the soil erodibility from the set of soil samples was carried out using a sequential Gaussian simulation. The results show that the simulation produces a reliable prediction map of soil erodibility and can be recommended as a monitoring strategy to spatially update soil erodibility.

Spatial uncertainty analysis for mapping soil erodibility based on joint sequential simulation

Soil erodibility (susceptibility of soil to be lost to erosion) is one of the components of the universal soil loss equation (USLE). In the USLE, erodibility is known as the K factor, which in turn is a function of these soil properties: particle size distribution, organic matter content, structure, and permeability. The traditional approach for estimating soil erodibility does not account for spatial variability of individual soil properties or spatial correlation among those properties. Our objectives in this study were to evaluate the use of joint sequential simulation for mapping soil erodibility, as well as to partition the individual and joint variance contribution of soil properties for predicting soil erodibility. We collected 192 usable soil samples across Fort Hood, Texas in the summer of 1999. For each of those samples, we obtained an estimate of particle size distribution, organic matter content, structure, permeability, and calculated soil erodibility. We carried out both independent and joint sequential simulation to generate spatially explicit predictions and variance of all soil properties as well as covariance between pairs of soil properties for each cell within our simulation area. We used the program GCOSIM3D to conduct those simulations. On average, joint sequential simulation resulted in a K factor variance of less than half the variance obtained from independent simulation. Using the results from joint sequential simulation, we partitioned the contribution of each soil property and pair of properties using first-order Taylor series expansion of the soil erodibility function. Individually, Very-Fine-Sand-and-Silt contributed the most (46.19%), whereas Structure contributed the least (6.53%) to the K factor variance. Jointly, Permeability/Structure contributed the most (9.32%), whereas Sand/Very-Fine-Sand-and-Silt caused the largest reduction (−19.19%) in the K factor variance. We conclude that joint sequential simulation provided approximately twice as much precision as independent simulation for the spatially explicit prediction of soil erodibility. Likewise, first-order Taylor series expansion offered an accurate approach for partitioning the individual and joint contribution of soil properties to soil erodibility variance. This partitioning allowed us to identify large sources of uncertainty and suggest efficient approaches for further improving the precision of K value predictions.

The calibration of digitized aerial photographs for forest stratification

Abstract The high spatial resolution of digitized aerial photographs may offer an accurate and effective means of mapping, inventorying, and monitoring forests. Due to the presence of bi-directional reflectance, however, the pixel values are affected by their location within the photo. Two similar sample plots or vegetation types in different parts of the photo may thus have quite dissimilar pixel values and texture features. It is consequently necessary to correct, or calibrate, pixel values when they are used in numerical interpretation. The effect of location of a window of pixels on various colour-infrared (CIR) aerial photographs corresponding to the field sample plots was analysed. Two calibration methods, regression calibration and ratioing, were derived and tested. Linear regression calibration to the principal-point level of the photos was shown to be the most effective, in which the mean pixel value of the window was modelled as a function of solar and sensor direction at the time of exposure. T...

Mapping and uncertainty of predictions based on multiple primary variables from joint co-simulation with Landsat TM image and polynomial regression

In the management of natural resources, multiple variables correlated with each other usually need to be mapped jointly. However, joint mapping and spatial uncertainty analyses are very difficult mainly because of interactions among variables and imperfection of existing methods. There is abundant evidence that considering interactions among variables and spatial information from neighbors can result in improved maps. This study presents a remote sensing-aided method for that purpose. The method is based on the integration of joint sequential co-simulation with Landsat TM image for mapping and polynomial regression for spatial uncertainty analysis. The method was applied to a case study in which ground cover (GC), canopy cover (CC), and vegetation height (VH) were jointly mapped to derive a map of the vegetation cover factor for predicting soil loss. The variance contributions from the variables, their interactions, and the spatial information from neighbors leading to uncertainty of predicted vegetation cover factor were assessed. The results showed that in addition to unbiased maps, this method reproduced the spatial variability of the variables and the spatial correlation among them, and successfully quantified the effect of variation from all the components on the prediction of the vegetation cover factor.

The Impact of Misclassification in Land Use Maps in the Prediction of Landscape Dynamics

Land use maps are widely used in modeling land use change, urban sprawl, and for other landscape related studies. A misclassification confusion matrix for land use maps is usually provided as a measure of their quality and uncertainty. However, this very important information is rarely considered in land use map based studies, especially in modeling landscape dynamics. Ignoring uncertainty of land use maps may cause models to provide unreliable predictions. This study is an attempt to investigate the impact of the accuracy of land use maps used as input for an urban sprawl model. In the study area, the regional confusion matrix has been localized using a topographical map. Based on the regional and local confusion matrices, several error levels have been defined. The results showed that a localized confusion matrix that reflected the characteristics of the study area had error rates that were much different than the regional confusion matrix. The predictions of the probability of urban sprawl based on the land use maps and defined error levels were quite different.

Improved generalized Fourier amplitude sensitivity test (FAST) for model assessment

The Fourier amplitude sensitivity test (FAST) can be used to calculate the relative variance contribution of model input parameters to the variance of predictions made with functional models. It is widely used in the analyses of complicated process modeling systems. This study provides an improved transformation procedure of the Fourier amplitude sensitivity test (FAST) for non-uniform distributions that can be used to represent the input parameters. Here it is proposed that the cumulative probability be used instead of probability density when transforming non-uniform distributions for FAST. This improvement will increase the accuracy of transformation by reducing errors, and makes the transformation more convenient to be used in practice. In an evaluation of the procedure, the improved procedure was demonstrated to have very high accuracy in comparison to the procedure that is currently widely in use.

A Methodology for Spatial Uncertainty Analysis Of Remote Sensing and GIS Products

When remote sensing and GIS products are generated, errors and uncertainties from collection, processing and analysis of image and ground data, and model development, accumulate and are propagated to the maps. The products thus possess many sources of uncertainties that vary spatially and temporally. Spatially identifying the sources of uncertainties, modeling their accumulation and propagation, and finally, quantifying them will be critical to control the quality of spatial data. This paper demonstrates a methodology and its applications for a case study in which uncertainty of predicted soil erosion is hierarchically partitioned into various primary components on a pixel-by-pixel basis. The methodology is based on a regionalized variable theory of variables. It integrates remote sensing aided co-simulation algorithms in geostatistics, and uncertainty and error budget methods in uncertainty analysis. The simulation algorithms generate realizations that can be used to calculate local estimates, and the variances and co-variances between them. Uncertainty and error budget methods partition the uncertainty of output into various input components and quantify their relative uncertainty contributions. The results can thus suggest the main uncertainty sources and their variation spatially, and further provide a rationale to reduce errors in map generation and application.