Journal of the Indian Society of Remote Sensing最新文献

The use of 3D Deep Learning (DL) models for LiDAR data segmentation has attracted much interest in recent years. However, the generation of labeled point cloud data, which is a prerequisite for training DL models, is a highly resource-intensive exercise. Simulated LiDAR data, which are already labeled, provide a cost-effective alternative, but their efficacy and usefulness must be evaluated. This paper examines the role of simulated LiDAR point clouds in training DL models. A high-fidelity 3D terrain model representing the real environment is developed, and the in-house physics-based simulator “Limulator” is used to generate labeled point clouds through various realizations. The paper outlines a few major hypotheses to assess the usefulness of simulated data in training DL models. The hypotheses are designed to assess the role of simulated data alone or in combination with real data or by strategic boosting of minor classes in simulated data. Several experiments are carried out to test these hypotheses. An experiment involves training a DL model, PointCNN in this case, using various combinations of simulated and real LiDAR data and measuring its performance to segment the test data. Results show that training using simulated data alone can produce an overall accuracy (OA) of 89% and the weighted-averaged F1 score of 88.81%. It is further observed that training using a combination of simulated and real data can achieve accuracies comparable to when only a large quantity of real data is employed. Strategic boosting of minor classes in simulated data improves the accuracies of minor classes by up to 23% compared to only real data. Training a DL model using simulated data, due to the ease in its generation and positive impact on segmentation accuracy, can be highly beneficial in the use of DL for LiDAR data. The use of simulated data for training has the potential to minimize the resource-intensive exercise of developing labeled real data.

To properly manage the terrestrial ecosystem, it is essential to understand the vegetation sensitivity to climate variations and human actions. The main target of this survey was to evaluate the spatiotemporal variation in vegetation cover, and its relationship with climate variations and to calculate the contributions of driving factors in Namak Lake basin, Iran, during 2001–2019. To this end, Vegetation Health Index (VHI) and Standardized Precipitation Evapotranspiration Index (SPEI) in 3, 6, 9, and 12-month time scales were used to assess vegetation dynamics and its reactions to climate variations based on coefficient of determination (R²) and Linear Regression (LR). The results presented that vegetation cover had an improving trend in 87.78% and a decreasing trend in 12.19% of the basin, while it was stable in 0.03% of areas. The correlation between VHI and different time scales of SPEI indicated that coverage was mainly affected by 3-month SPEI in more than half of the basin (53.74%). High correlations between VHI and SPEI were found in upland areas in the northeast and some areas in the east of the basin. These areas also had the highest slope of VHI changes in relation to climate factors. Climate variability affected about four-fifths (79.22%) of coverage, while 16.36% was influenced by human actions, and 4.42% by both factors. Moreover, more than 99% of the significant improvements and degradations in coverage were related to climate variations and mankind’s actions, respectively. The outcomes can serve as a foundation for initiating vegetation growth and protection in the Namak Lake basin.

As an important forest type, deciduous broad-leaved forest is crucial for estimating forest carbon sequestration capacity and evaluating forest carbon balance. This study focuses on the natural deciduous broad-leaved forest of Mazongling Nature Reserve in Jinzhai County of China. WorldView-2 images were selected as data source. 36 candidate factors including vegetation indices, texture features, and topographic factors were used for modelling. Three machine learning algorithms (i.e., random forest, k-nearest neighbor, and artificial neural network) were used to establish the optimal quantitative retrieval model for natural deciduous broad-leaved biomass. Results showed that the ANN model was the best predictor with R² = 0.69 and RMSE = 31.53 (Mg·ha⁻¹). Combining the ANN model with the complete spatial coverage of remote sensing data, we developed a distribution map of natural deciduous broad-leaved biomass in the Mazongling forest farm. The estimated average biomass of the study area was 90.34 ± 47.96 Mg·ha⁻¹. In addition, the influence of light saturation on model accuracy is also discussed. This study confirms that remote sensing data in temporal and spatial space can improve the model estimation accuracy.

The pattern of monsoon rainfall has been explored at diurnal scale level using the precipitation dataset GPM-3IMERGHH-06 over the state of Himachal Pradesh from 2004 to 2020. The present study identified the spatio-temporal pattern of rainfall peaks and found the trend of rainfall peaks (5% significance level) during different times over selected locations. One-sample student’s t-test has been employed to find the significance of the trend. Intense rainfall peaks (0.7–1 mm/hour) are found at late night (0200 IST to 0230 IST) as well as morning (0600 IST TO 0630 IST) hours over the mountain foot regions and less intense peaks (0.4–0.6 mm/hour) are found during afternoon or evening time (1430 IST to 1700 IST) over southern parts of the state. Late-night peaks of rainfall show a significant increasing trend over the regions Kullu, Mandi, and Sirmaur, while morning peaks of rainfall show a decreasing trend over the same regions. Significant increasing trends have been found over Hamirpur and Mandi during evening hours. The state is characterized by undulating terrain and is prone to extreme rainfall events during the monsoon season. Diurnal pattern of rainfall gives a glimpse into the physical mechanism behind such sudden unexpected events. Trend analysis helps to understand the future risk over different regions which is important for implementing any kind of mitigative actions.

Measuring atmospheric winds over longer ranges using VHF-MST radar is extremely useful for studying stratosphere-troposphere exchange. The present study uses an adaptive block technique (ABlockJS), a mixed model of a parametric technique, and a few non-parametric techniques to address this aspect. The signal estimates are substantiated with five moments and six quality-related parameters while deriving three wind components along with horizontal wind speed and direction. The parametric part of the technique improves the signal, while the non-parametric part lowers noise variance. This technique is established using NARL MST Radar experimental data. The computed wind components derived from this technique are verified with the independent wind components acquired from the concurrent GPS radiosonde in-situ observations. It is observed that this analytical technique can deliver wind components more precisely and consistently, covering longer ranges of 25.20 km. It enhances the benchmark range coverage of 21.45 km attained using Fourier-based estimators on the MST dataset. The complete procedure is developed in C# from scratch without using any standard routine from available packages, thus, it fits acquisition-time application needs fine. It benefits various atmospheric research which demands higher range coverage using VHF radar.

This study integrates Remote Sensing data, field investigation, and petrography to analyze the Zawar Pb–Zn sulfide deposits, in the Paleoproterozoic Aravalli Supergroup rocks of NW India. Structural features of the study area are delineated using Remote Sensing and Shuttle Radar Topography Mission (SRTM) data. Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) data, is used to distinguish the major rock types and alteration zones. Our findings reveal that the Zawar belt is composed of phyllite, quartzite, carbonate, and, greywacke. Phyllites from the hinge of the Main Zawar Fold (MZF) provide critical insights into the distribution of monazite veins, and, support the evidence of hydrothermal alteration over the hinge area of the MZF. Textural evidence investigated by Scanning Electron Microscopic study (SEM) suggests that the monazite is of epigenetic hydrothermal origin, formed subsequently after the formation of the primary host rock. Energy Dispersive X-ray spectroscopic (EDS) study indicates that these monazites have an average composition, P₂O₅(17.85 wt.%), Ce₂O₃ 14.49, La₂O₃ 6.98, Nd₂O₃ 5.39 and ThO₂ 1.60 wt.%, suggesting its hydrothermal origin.

Drought is considered to be the most complex but least understood of all natural hazards, affecting more people. Its reappearance in drought-prone areas every few years is almost certain. Also, they lack sudden and easily identified onsets and terminations. Under the background of global climate change, the impact from drought exhibits the characteristics of complexity and multi-process. It has a significant impact on the water resources, agriculture, society, and economy hence needs attention. Vegetation Condition Index (VCI) is used for observing the change in vegetation that causes agricultural drought. Since the land surface temperature has minimum influence from cloud contamination and humidity in the air, so the Temperature Condition Index (TCI) is used for studying the temperature change. Dryness or wetness of soil is a major indicator for agriculture and a comprehensive assessment of vegetation and temperature stress is achieved from MODIS satellite data in Google Earth Engine (GEE) platform for pre and post monsoon season from 2008 to 2022 (15- year period). Vegetation Condition Index (VCI) is used for observing the change in vegetation that causes agricultural drought. Since the land surface temperature has minimum influence from cloud contamination and humidity in the air, so the Temperature Condition Index (TCI) is used for studying the temperature change. The research also incorporates precipitation data from WorldClim to investigate its influence on the Vegetation Health Index (VHI). Mann Kendall trend analysis is employed to examine spatio-temporal variations in drought severity, for both pre-monsoon and post-monsoon seasons. The results emphasize the sensitivity of VHI to shifts in rainfall patterns, providing valuable insights for drought monitoring and management. In essence, this study enhances understanding of drought dynamics and emphasizes the significance of Remote Sensing data and climate information for effective drought assessment and mitigation strategies.

The genus Heterobasidion includes some of the most destructive pathogens of conifers in the Northern hemisphere. Heterobasidion root rot leads to loss of root function and visible symptoms in the crowns of most Pinus spp., including Turkish red pine (P. brutia). Infected pines will eventually die. Wind-thrown trees with decayed roots or open gaps in the stand often indicate the presence of Heterobasidion root rot. Satellite imagery has recently been utilized regularly to detect damaged areas in order to apply early management procedures to pests or diseases in forests, reducing spread within an affected site and to other places. In the work described here, Sentinel-2 A satellite imagery was tested for detecting Heterobasidion root rot in P. brutia regeneration in an area in south-western Turkiye, using different vegetation indices. Normalized Difference Red Edge Index (NDRE), Normalized Difference Vegetation Index (NDVI), and Plant Senescence Reflectance Index (PSRI) indices were calculated from Sentinel-2 A satellite images in the Google Earth Engine (GEE) platform to detect disease. Calculated indices as synthetic band were added to the Sentinel-2 A satellite image on the GEE platform. Images with the added bands were classified using Random Forest (RF) before evaluation using the Kappa Coefficient and Overall Accuracy. Based on a statistical analysis, NDRE was the most useful index for detecting the disease with an overall accuracy of 89% and a Kappa Coefficient of 0.84, followed by NDVI and PSRI, respectively. After evaluation of General Accuracy and Kappa Coefficient, disease incidence in the area was determined (affected hectares), based on the indices. NDRE detected 7.21 affected hectares, NDVI 7.9 hectares and PSRI 6.49 hectares in a total of 67.8 hectares. Sentinel-2 A bands, which allow the measurement of various land and vegetation health parameters, the effect of bands on RF classification was determined according to the indices used. The most important band for classification of NDRE and NDVI was the B2 (BLUE) band of Sentinel-2 A, and the most important band with PSRI was the B5 (RED EDGE) band. Based on these bands, the best wavelengths for detecting H. annosum diseased areas were in the range 492.4–740.5 nm in Sentinel-2 A. The system enabled the detection of differences in crown deterioration and also wind-thrown trees with decayed roots or open gaps in the stand. This study is the first to show that Sentinel-2 A satellite imagery can be applied successfully for the detection of Heterobasidion root rot on P. brutia.

Methane (CH₄) is a potent greenhouse gas and the second highest anthropogenic emissions are recorded from CH₄ on Earth. Considering its high global warming potential, the monitoring of source locations is inadvertent. The paper presented here is the first attempt (to the best of our knowledge) to comprehensively analyse the methane emissions over multiple Indian locations using satellite data. It outlays a brief background of methane emission sensors and studies carried out worldwide for estimation of the GHG. It further enumerates the potential of Earth Surface Mineral Dust Source Investigation (EMIT) and TROPOspheric Monitoring Instrument (TROPOMI) in highlighting the potential point sources of methane emissions and its concentration/emission flux in India. 17 unique plumes were identified using EMIT in the states of Maharashtra (06), Rajasthan (04), Punjab (02), Gujarat (03) and Assam (02). Gujarat, Surat, Assam Uttar Pradesh and Haryana using TROPOMI were also studied. The hotspots showcase emission sources from solid waste landfill sites, sewage treatment plants, wetlands/marshy agriculture, city sewage outlets, oil and gas fields, oil refinery and textile industry. It was observed that EMIT can effectively be used for point source identification, monitoring and enhancement while TROPOMI is best suited for regional level methane monitoring. A sewage outlet plume in Maharashtra produced the maximum emission of 6202.9 ± 691.94 kg/hr followed by solid waste (SW) sites located in Pirana Landfill, Ahmedabad and Khajod Landfill, Surat in Gujarat. Methane monitoring is an important step towards mitigating enormous methane emissions and anomalous methane sources.

Snow has a significant impact on the hydrological cycle, contributing to energy generation, meeting agricultural demands, and providing drinking water. Effective management of snowmelt runoff can help control and prevent potential risks. The purpose of the study is to evaluate the use of the remotely sensing data to improve the estimation accuracy of the snowmelt-runoff by using the Snowmelt-Runoff Model (SRM). To do this, a total of 1595 Tropical Rainfall Measuring Mission (TRMM) and Moderate Resolution Imaging Spectroradiometer (MODIS) satellite images were prepared between 2014 and 2015 to acquire data on precipitation, minimum and maximum temperatures and Snow Cover Area (SCA). The accuracy of precipitation data was evaluated using Root Mean Squared Error (RMSE) and Root Mean Squared Log-Error (RMSLE) to ensure their reliability. Additionally, a sensitivity analysis of the SRM model’s coefficients, particularly for recession coefficient (K) and snow runoff (C_s), was conducted to understand their impact on the model’s performance. In this study, meteorological station data and satellite data from the years 2014 and 2015 were utilized for the validation and calibration stages, respectively. The model’s ability to estimate snowmelt runoff using remote sensing data was evaluated using both on-site stations and satellite data. In the calibration period, the snowmelt runoff estimation results were obtained with Nash-Sutcliffe Efficiency (NSE) index values of 0.72 and 0.70 for on-site stations and satellite data, respectively. In the validation period, the NSE index values were 0.60 and 0.93 for on-site stations and satellite data, respectively indicating improved performance when using satellite data to estimate the snowmelt runoff. The study’s findings show that remote sensing data enhances the performance of the SRM model for estimating the snowmelt-runoff.