Search

Intensive use of military vehicles in military installations create conditions favorable for gully formation. Gullies impede the access of vehicle, restrict the continuation of training, and lead to significant damage to vehicle and risk the life of soldiers. Therefore, it is critical to correctly identify the locations of gullies for continuous training mission. In this study, Fort Riley (FR) military installation was chosen as the study area. LiDAR derived 1 m resolution digital elevation model (DEM) acquired on 2010 was used to map the gullies. A procedure that measures local topographic position, i.e., difference from mean elevation (DFME) along with its integration to the land surface having high surface curvature values was employed. Two high spatial resolution WorldView-2 images of 2010 and field gully data collected in 2010 were utilized for accuracy assessment. Results showed that: (1) A total of 237 small and 166 large gullies were detected and most of them dominated the central west and northwest parts of the installation; (2) Based on the visual interpretation in the WorldView-2 images, there was no statistically significant difference between the detected and observed numbers of gullies; (3) Gullies measured in the field were well detected with an overall accuracy of 78%. [ABSTRACT FROM AUTHOR]

Land degradation and desertification in arid and semi-arid areas is of great concern. Accurately mapping percentage vegetation cover (PVC) of the areas is critical but challenging because the areas are often remote, sparsely vegetated, and rarely populated, and it is difficult to collect field observations of PVC. Traditional methods such as regression modeling cannot provide accurate predictions of PVC in the areas. Nonparametric constant k-nearest neighbors (Cons_kNN) has been widely used in estimation of forest parameters and is a good alternative because of its flexibility. However, using a globally constant k value in Cons_kNN limits its ability of increasing prediction accuracy because the spatial variability of PVC in the areas leads to spatially variable k values. In this study, a novel method that spatially optimizes determining the spatially variable k values of Cons_kNN, denoted with Opt_kNN, was proposed to map the PVC in both Duolun and Kangbao County located in Inner Mongolia and Hebei Province of China, respectively, using Landsat 8 images and sample plot data. The Opt_kNN was compared with Cons_kNN, a linear stepwise regression (LSR), a geographically weighted regression (GWR), and random forests (RF) to improve the mapping for the study areas. The results showed that (1) most of the red and near infrared band relevant vegetation indices derived from the Landsat 8 images had significant contributions to improving the mapping accuracy; (2) compared with LSR, GWR, RF and Cons_kNN, Opt_kNN resulted in consistently higher prediction accuracies of PVC and decreased relative root mean square errors by 5%, 11%, 5%, and 3%, respectively, for Duolun, and 12%, 1%, 23%, and 9%, respectively, for Kangbao. The Opt_kNN also led to spatially variable and locally optimal k values, which made it possible to automatically and locally optimize k values; and (3) the RF that has become very popular in recent years did not perform the predictions better than the Opt_kNN for the both areas. Thus, the proposed method is very promising to improve mapping the PVC in the arid and semi-arid areas. [ABSTRACT FROM AUTHOR]

The degradation of permafrost in the Earth's cold regions due to climate warming and human activities is of increasing concern because of its impact on ecosystems and land management, thus, monitoring the degradation becomes critical. This study aimed to demonstrate a method in which Sentinel-1 DInSAR and MODIS data were combined to estimate and map seasonal and long-term land surface deformation (subsidence/uplifting) and permafrost active layer thickness (ALT) in Donnelly Training Area (DTA) in Alaska. The coherence (similarity) between the paired SAR images was analyzed for their applicability. A sensitivity analysis and accuracy assessment of the estimates were conducted to account for their quality. Results showed that the estimated seasonal subsidence mainly occurring in June and July was in the range of 0 to −0.43 m, while the estimated uplifting mainly happening from September to May of next year was as great as +0.34 m. Moreover, the estimated long-term (2015 to 2018) subsidence and uplifting were mostly distributed in the southern and northern parts of DTA, respectively. The spatially variable time delays led to the ALT estimates greater along the east river and in the west and south parts, and smaller in the north parts. At the significance level of 0.05, the coherence estimates of the paired images were significantly different from zero and compared with the referenced predictions from a widely used annual prediction model, the average residual of the ALT estimates did not significantly differ from zero at the significance level of 0.05. The spatial distributions of the uncertainties for the seasonal surface deformation estimates were similar to those of the input uncertainties from the estimates of model coefficients, the phase change, the modelling error, and the image pairs. Thus, the DInSAR image based method coupled with MODIS data offered the potential of mapping and monitoring the dynamics of permafrost environment for the cold regions such as DTA in which collecting field observations is difficult and costly. This study also enhanced understanding spatiotemporal variability of permafrost deformation in DTA and provided guidelines for developing a near real-time monitoring system of the permafrost environment. • Permafrost degradation due to climate warming and human activity is of great concern. • Land surface deformation of an installation in Alaska is mapped by SAR&MODIS images. • A quality assessment of the land surface deformation estimates is conducted. • The accuracy of the estimates is compatible with that from a widely used model. • This study provides guidelines for developing a near real-time monitoring system. [ABSTRACT FROM AUTHOR]