Multi-sensor Radar Research Articles

Estimation of ground subsidence of New Delhi, India using PS-InSAR technique and Multi-sensor Radar data

Globally the land subsidence is a significant problem of the rapidly growing urban area. The factor responsible for the land subsidence caused by over-exploitation of the underground fluid such as water, petroleum, and gas respectively. In present study we present the result of detail investigation of active ground subsidance in New Delhi, National Capital Region (NCR). This area indicates a high rate of urban growth during the past decades. To analyze the land subsidence, we used multiple SAR sensor data and exploited the PS-InSAR technique. The data used for this study are Cosmo-skymed acquired between 08/06/2011 to 15/11/2017, Sentinel-1A-B (18-12-2014 to 27-11-2018), and ALOS PALSAR acquired between 19/01/2007 and 20/01/2011. These radar sensors operate in X, C, and L-band, which covers over ten years, from 2007 to 2018. The PSI results of Cosmo-skymed reveals that the Delhi NCR region has undergone an average deformation ± 15 mm/y, a maximum surface deformation observed from ALOS-PALSAR is 10 to 18 mm/y and the ground displacement observed from the SENTINAL-1A data is −2 to 16 mm/y. Groundwater level data also collected for the same period and a ground water level depletion compared with the subsidence. Monitoring land subsidence with ground-based conventional technology is time-consuming and can be carried out in a limited area due to the financial implication. PS-InSAR is an established method to detect the surface movement using the SAR sensor's time-series data. The result shows that a twenty centimeter of land subsidence is visible in some areas, validated with the collected ground evidence. The affected area is also showing resemblance to the groundwater depleting condition in those areas. This study also establish that multiple sensor data can be used to monitor the long term land subsidence.

Mapping Mountain Peatlands and Wet Meadows Using Multi-Date, Multi-Sensor Remote Sensing in the Cordillera Blanca, Peru

Wetlands (called bofedales in the Andes of Peru) are abundant and important components of many mountain ecosystems across the globe. They provide many benefits including water storage, high quality habitat, pasture, nutrient sinks and transformations, and carbon storage. The remote and rugged setting of mountain wetlands creates challenges for mapping, typically leading to misclassification and underestimates of wetland extent. We used multi-date, multi-sensor radar and optical imagery (Landsat TM/PALSAR/RADARSAT-1/SRTM DEM-TPI) combined with ground truthing for mapping wetlands in Huascaran National Park, Peru. We mapped bofedales into major wetland types: 1) cushion plant peatlands, 2) cushion plant wet meadows, and 3) graminoid wet meadows with an overall accuracy of 92%. A fourth wetland type was found (graminoid peatlands) but was too rare to map accurately, thus it was combined with cushion peatland to form a single peatland class. Total wetland area mapped in the National Park is 38,444 ha, which is 11% of the park area. Peatlands were the most abundant wetland type occupying 6.3% of the park, followed by graminoid wet meadows (3.5%) and cushion wet meadows (1.3%). These maps will serve as the foundation for improved management, including restoration, and estimates of landscape carbon stocks.

Sparse representation discretization errors in multi-sensor radar target motion estimation

Abstract. In a multi-sensor radar for the estimation of the targets motion states, more than one module of transmitter and receiver are utilized to estimate the positions and velocities of targets, also known as motion states. By applying the compressed sensing (CS) reconstruction algorithms, the surveillance space needs to be discretized. The effect of the additive errors due to the discretization are studied in this paper. The errors are considered as an additive noise in the well-known under-determined CS problem. By employing properties of these errors, analytical models for its average and variance are derived. Numerous simulations are carried out to verify the analytical model empirically. Furthermore, the probability density functions of discretization errors are estimated. The analytical model is useful for the optimization of the performance, the efficiency and the success rate in CS reconstruction for radar as well as many other applications.

Multidate, multisensor remote sensing reveals high density of carbon-rich mountain peatlands in the páramo of Ecuador.

Tropical peatlands store a significant portion of the global soil carbon (C) pool. However, tropical mountain peatlands contain extensive peat soils that have yet to be mapped or included in global C estimates. This lack of data hinders our ability to inform policy and apply sustainable management practices to these peatlands that are experiencing unprecedented high rates of land use and land cover change. Rapid large-scale mapping activities are urgently needed to quantify tropical wetland extent and rate of degradation. We tested a combination of multidate, multisensor radar and optical imagery (Landsat TM/PALSAR/RADARSAT-1/TPI image stack) for detecting peatlands in a 2715km2 area in the high elevation mountains of the Ecuadorian páramo. The map was combined with an extensive soil coring data set to produce the first estimate of regional peatland soil C storage in the páramo. Our map displayed a high coverage of peatlands (614km2 ) containing an estimated 128.2±9.1 Tg of peatland belowground soil C within the mapping area. Scaling-up to the country level, páramo peatlands likely represent less than 1% of the total land area of Ecuador but could contain as much as ~23% of the above- and belowground vegetation C stocks in Ecuadorian forests. These mapping approaches provide an essential methodological improvement applicable to mountain peatlands across the globe, facilitating mapping efforts in support of effective policy and sustainable management, including national and global C accounting and C management efforts.

Assessment of multi-temporal, multi-sensor radar and ancillary spatial data for grasslands monitoring in Ireland using machine learning approaches

Accurate inventories of grasslands are important for studies of carbon dynamics, biodiversity conservation and agricultural management. For regions with persistent cloud cover the use of multi-temporal synthetic aperture radar (SAR) data provides an attractive solution for generating up-to-date inventories of grasslands. This is even more appealing considering the data that will be available from upcoming missions such as Sentinel-1 and ALOS-2. In this study, the performance of three machine learning algorithms; Random Forests (RF), Support Vector Machines (SVM) and the relatively underused Extremely Randomised Trees (ERT) is evaluated for discriminating between grassland types over two large heterogeneous areas of Ireland using multi-temporal, multi-sensor radar and ancillary spatial datasets. A detailed accuracy assessment shows the efficacy of the three algorithms to classify different types of grasslands. Overall accuracies≥88.7% (with kappa coefficient of 0.87) were achieved for the single frequency classifications and maximum accuracies of 97.9% (kappa coefficient of 0.98) for the combined frequency classifications. For most datasets, the ERT classifier outperforms SVM and RF.