Assessment and estimation of coal dust impact on vegetation using VIs difference model and PRISMA hyperspectral data in mining sites.

This work focuses on dust detection, and estimation of vegetation in coal mining sites using the vegetation indices (VIs) differences model and PRISMA hyperspectral imagery. The results were validated by ground survey spectral and foliar dust data. The findings indicate that the highest Separability (S), Coefficient of discrimination (R2), and lowest Probability (P) values were found for the narrow-banded Narrow-banded Normalized Difference Vegetation Index (NDVI), Transformed Soil Adjusted Vegetation Index (TSAVI), and Tasselled Cap Transformation Greenness (TC-greenness) indices. These indices have been utilized for the Vegetation Combination (VC) index analysis. Compared to other VC indices, this VC index revealed the highest difference (29.77%), which led us to employ this index for the detection of healthy and dust-affected areas. The foliar dust model was developed for the estimation and mapping of dust impact on vegetation using the VIs differences models (VIs diff models), laboratory dust amounts, and leaf spectral regression analysis. Based on the highest R2 (0.90), the narrow-banded TC-greenness differenced VI was chosen as the best VI, and the coefficient (L) value (-7.75gm/m2) was used for estimating the amount of foliar dust in coal mining sites. Compared to other indices-based difference dust models, the narrow-banded TC-greenness difference image had the highest R2 (0.71) and lowest RMSE (4.95 gm/m2). According to the findings, the areas with the highest dust include those with mining haul roads, transportation, rail lines, dump areas, tailing ponds, backfilling, and coal stockyard sides. This study also showed a significant inverse relationship (R2 = 0.84) among vegetation dust classes, leaf canopy spectrum, and distance from mines. This study provides a new way for estimating dust on vegetation based on advanced hyperspectral remote sensing (PRISMA) and field spectral analysis techniques that may be helpful for vegetation dust monitoring and environmental management in mining sites.

Vegetation health conditions assessment and mapping using AVIRIS-NG hyperspectral and field spectroscopy data for -environmental impact assessment in coal mining sites.

This paper focuses on vegetation health conditions (VHC) assessment and mapping using high resolution airborne hyperspectral AVIRIS-NG imagery and validated with field spectroscopy-based vegetation spectral data. It also quantified the effect of mining on vegetation health for geo-environmental impact assessment at a fine level scale. In this study, we have developed and modified vegetation indices (VIs) based model for VHC assessment and mapping in coal mining sites. We have used thirty narrow banded VIs based on the statistical measurement for suitable VIs identification. The highest Pearson's r, R2, lowest RMSE, and P values indices have been used for VIs combined pixels analysis. The highest different (Healthy vs. unhealthy) vegetation combination index (VCI) has been selected for VHC assessment and mapping. We have also compared VIs model-based VHC results to ENVI (software) forest health tool and Spectral-based SAM classification results. The 1st VCI result showed the highest difference (72.07%) from other VCI. The AUC values of the ROC curve have shown a better fit for the VIs model (0.79) than Spectral classification (0.74), and ENVI FHT (0.68) based on VHC results. The VHC results showed that unhealthy vegetation classes are located at low distances from mine sites, and healthy vegetation classes are situated at high distances. It is also seen that there is a highly significant positive relationship (R2 =0.70) between VHC classes and distance from mines. These results will provide a guideline for geo-environmental impact assessment in coal mining sites.

Deforestation susceptibility assessment and prediction in hilltop mining-affected forest region.

This work mainly focused on deforestation susceptibility (DS) assessment and its prediction based on statistical models (FR, LR & AHP) in the Saranda forest, India. Also, efforts had been made to quantify the effect of mining on deforestation. We had considered twenty-five (twenty present and five predicted) causative variables of deforestation, including climate, natural or geomorphological, forestry, topographical, environmental, and anthropogenic. The predicted variables have been generated from different simulation models. Also, very high-resolution, Google Earth imagery have been used in time series analysis for deforestation from 1987 to 2020 data and generated dependent variable. On deforestation analysis, it was observed that a total of 4197.84 ha forest areas were lost in the study region due to illegal mining, agricultural and tribal people allied activities. The DS results have shown that of total existing forest area, 11.22% area were under very high, 16.08% under high, 16.18% under moderate, 24.25% under low, and 32.27% falls very low categories. According to the DS assessment and predicted results, the very high susceptibility classes were found at and close to mines, agricultural, roads and settlement's surrounding sites. The sensitivity analysis results also shown that some causative variables (maximum temperature (2.95%), minimum temperature (0.51%), rainfall (2.69%), LST (4.56%), hot spot (7.36%), aspect (1.14%), NDVI (2.64%), forest density (3.78%), lithology (3.26%), geomorphology (3.00%), distance from agricultural (19.40%), soil type (2.05%), solar radiation (5.97%), LULC (3.26%), drought (3.16%), altitude (2.85%), slope (5.97%), distance from mines (18.05%), roads (2.17%), and settlements (5.18%)) were more sensitive to deforestation. Most of the sensitive parameters showed a positive correlation with DS. The AUC values of the ROC curve had shown a better fit for AHP (0.72) than (0.69) FR and LR (0.68) models for present DS results. The correlation results had shown a good inverse relationship between DS and distance from mines and foliar dust concentration. This work will espouse the future work in the effective planning and management of the mining-affected forest region and predicted deforestation susceptibility would be helpful for forest ecosystem study and policymaking.

Assessment of mining activities on tree species and diversity in hilltop mining areas using Hyperion and Landsat data.

The tree species and its diversity are two critical components to be monitored for sustainable management of forest as well as biodiversity conservation. In the present study, we have classified the tree species and estimated its diversity based on hyperspectral remote sensing data at a fine scale level in the Saranda forest. This area is situated near the mining fields and has a dense forest cover around it. The forest surrounding the study area is exhibiting high-stress condition as evidenced by the dying and dry plant material, consequently affecting tree species and its diversity. The preprocessing of 242 Hyperion (hyperspectral) spectral wavebands resulted in 145 corrected spectral wavebands. The 21 spectral wavebands were selected through discrimination analysis (Walk's Lambda test) for tree species analysis. The SVM (support vector machine), SAM (spectral angle mapper), and MD (minimum distance) algorithms were applied for tree species classification based on ground spectral data obtained from the spectroradiometer. We have identified six local tree species in the study area at the spatial level. The result shows that Sal and Teak tree species are located in the upper and lower hilly sides of two mines (Meghahatuburu and Kiriburu). We have also used hyperspectral narrow banded vegetation indices (VIs) for species diversity estimation based on the field-measured Shannon diversity index. The statistical result shows that NDVI705 (red edge normalized difference vegetation index) is having the best R2 (0.76) and lowest RMSE (0.04) for species diversity estimation. That is why we have used NDVI705 for species diversity estimation. The result shows that higher species diversity values are located in the upper and lower hilly sides of two mines. The linear regression between Hyperion and field measured Shannon index shows the R2 (0.72) and RMSE (0.15). This study will aid in effective geoenvironmental planning and management of forest in the hilltop mining areas.