Dataset Information

Predicting active-layer soil thickness using topographic variables at a small watershed scale.

ABSTRACT: Knowledge about the spatial distribution of active-layer (AL) soil thickness is indispensable for ecological modeling, precision agriculture, and land resource management. However, it is difficult to obtain the details on AL soil thickness by using conventional soil survey method. In this research, the objective is to investigate the possibility and accuracy of mapping the spatial distribution of AL soil thickness through random forest (RF) model by using terrain variables at a small watershed scale. A total of 1113 soil samples collected from the slope fields were randomly divided into calibration (770 soil samples) and validation (343 soil samples) sets. Seven terrain variables including elevation, aspect, relative slope position, valley depth, flow path length, slope height, and topographic wetness index were derived from a digital elevation map (30 m). The RF model was compared with multiple linear regression (MLR), geographically weighted regression (GWR) and support vector machines (SVM) approaches based on the validation set. Model performance was evaluated by precision criteria of mean error (ME), mean absolute error (MAE), root mean square error (RMSE), and coefficient of determination (R2). Comparative results showed that RF outperformed MLR, GWR and SVM models. The RF gave better values of ME (0.39 cm), MAE (7.09 cm), and RMSE (10.85 cm) and higher R2 (62%). The sensitivity analysis demonstrated that the DEM had less uncertainty than the AL soil thickness. The outcome of the RF model indicated that elevation, flow path length and valley depth were the most important factors affecting the AL soil thickness variability across the watershed. These results demonstrated the RF model is a promising method for predicting spatial distribution of AL soil thickness using terrain parameters.

SUBMITTER: Li A

PROVIDER: S-EPMC5587236 | biostudies-literature | 2017

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Predicting active-layer soil thickness using topographic variables at a small watershed scale.

Li Aidi A Tan Xing X Wu Wei W Liu Hongbin H Zhu Jie J

PloS one 20170906 9

Knowledge about the spatial distribution of active-layer (AL) soil thickness is indispensable for ecological modeling, precision agriculture, and land resource management. However, it is difficult to obtain the details on AL soil thickness by using conventional soil survey method. In this research, the objective is to investigate the possibility and accuracy of mapping the spatial distribution of AL soil thickness through random forest (RF) model by using terrain variables at a small watershed s ...[more]

PMID: 28877196

Similar Datasets

Project description:Carbon release from thawing permafrost soils could significantly exacerbate global warming as the active-layer deepens, exposing more carbon to decay. Plant community and soil properties provide a major control on this by influencing the maximum depth of thaw each summer (active-layer thickness; ALT), but a quantitative understanding of the relative importance of plant and soil characteristics, and their interactions in determine ALTs, is currently lacking. To address this, we undertook an extensive survey of multiple vegetation and edaphic characteristics and ALTs across multiple plots in four field sites within boreal forest in the discontinuous permafrost zone (NWT, Canada). Our sites included mature black spruce, burned black spruce and paper birch, allowing us to determine vegetation and edaphic drivers that emerge as the most important and broadly applicable across these key vegetation and disturbance gradients, as well as providing insight into site-specific differences. Across sites, the most important vegetation characteristics limiting thaw (shallower ALTs) were tree leaf area index (LAI), moss layer thickness and understory LAI in that order. Thicker soil organic layers also reduced ALTs, though were less influential than moss thickness. Surface moisture (0-6 cm) promoted increased ALTs, whereas deeper soil moisture (11-16 cm) acted to modify the impact of the vegetation, in particular increasing the importance of understory or tree canopy shading in reducing thaw. These direct and indirect effects of moisture indicate that future changes in precipitation and evapotranspiration may have large influences on ALTs. Our work also suggests that forest fires cause greater ALTs by simultaneously decreasing multiple ecosystem characteristics which otherwise protect permafrost. Given that vegetation and edaphic characteristics have such clear and large influences on ALTs, our data provide a key benchmark against which to evaluate process models used to predict future impacts of climate warming on permafrost degradation and subsequent feedback to climate.

Project description:Climate change can have particularly severe consequences for high-elevation species that are well-adapted to long-lasting snow conditions within their habitats. One such species is the wolverine, Gulo gulo, with several studies showing a strong, year-round association of the species with the area defined by persistent spring snow cover. This bioclimatic niche also predicts successful dispersal paths for wolverines in the contiguous United States, where the species shows low levels of genetic exchange and low effective population size. Here, we assess the influence of additional climatic, vegetative, topographic, and anthropogenic, variables on wolverine genetic structure in this region using a multivariate, multiscale, landscape genetic approach. This approach allows us to detect landscape-genetic relationships both due to typical, small-scale genetic exchange within habitat, as well as exceptional, long-distance dispersal among habitats. Results suggest that a combination of snow depth, terrain ruggedness, and housing density, best predict gene flow in wolverines, and that the relative importance of variables is scale-dependent. Environmental variables (i.e., isolation-by-resistance, IBR) were responsible for 79% of the explained variation at small scales (i.e., up to ~230 km), and 65% at broad scales (i.e., beyond ~420 km). In contrast, a null model based on only space (i.e., isolation-by-distance, IBD) accounted only for 17% and 11% of the variation at small and broad scales, respectively. Snow depth was the most important variable for predicting genetic structures overall, and at small scales, where it contributed 43% to the variance explained. At broad spatial scales, housing density and terrain ruggedness were most important with contributions to explained variation of 55% and 25%, respectively. While the small-scale analysis most likely captures gene flow within typical wolverine habitat complexes, the broad-scale analysis reflects long-distance dispersal across areas not typically inhabited by wolverines. These findings help to refine our understanding of the processes shaping wolverine genetic structure, which is important for maintaining and improving functional connectivity among remaining wolverine populations.

Project description:BackgroundSoil erosion is a severe problem in the karst watershed, and analysis of soil erosion at the watershed scale is urgently needed.MethodsThis study tried to estimate the soil erodibility factor (K-factor) using the Erosion Productivity Impact Calculator (EPIC) nomograph and evaluate the spatial distribution of the predicted K-factor in a karst watershed. Soil properties and K-factors of five land use types (NF: natural mixed forest, CF: cypress forest, EF: economic forest, ST: stone dike terrace, VF: vegetable land) in the Xialaoxi small watershed were compared and key factors affecting erodibility were analyzed.ResultsResults showed that (1) The erodibility K-factor was unevenly distributed within different site types and strongly influenced by anthropogenic activities. The soil K-factors of sample sites subjected to frequent human disturbance (ST, VF) were high, ranging from 0.0480-0.0520 t hm2 h/(MJ mm hm2), while the soil K-factors of natural site types (NF, CF, and EF) were low, ranging from 0.0436-0.0448 t hm2 h/(MJ mm hm2). (2) The soil texture in the Xialaoxi watershed was mostly loamy, and that of the agricultural areas frequently disturbed by agricultural practices (ST, VF) was silty loam. (3) Soil carbon fractions were affected by land use types. Soil organic carbon storage of NF and CF had strong spatial heterogeneity. The soil organic carbon (SOC) and labile organic carbon (LOC) of the two were significantly higher than those of the disturbed EF and cultivated land soil. (4) There was a synergistic effect between the soil properties and the K-factor. K was significantly negatively related to sand fractions (2-0.05 mm) and non-capillary porosity, while positively related to silt content (0.05-0.002 mm). Overall, changes in bulk density (BD), total porosity (TP), non-capillary porosity (NCP), texture, and organic matter content caused by natural restoration or anthropogenic disturbance were the main reasons for soil erodibility. Natural care (sealing) and construction of stone dike planting practices were effective ways to reduce soil erosion in small karst watershed areas of western Hubei.

Dataset Information

Predicting active-layer soil thickness using topographic variables at a small watershed scale.

Publications

Predicting active-layer soil thickness using topographic variables at a small watershed scale.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets