The accuracy of land crop maps obtained from satellite images depends on the type of feature selection algorithm and classifier. Each of these algorithms have different efficiency in different ...conditions; therefore, developing a suitable strategy for combining the capabilities of different algorithms in preparing a land crop map with higher accuracy can be very useful. The objective of this study was to develop a fusion-based framework for improving land crop mapping accuracy. First, the features were retrieved using the Sentinel 1, Sentinel 2, and Landsat-8 imagery. Then, training data and various feature selection algorithms including recursive feature elimination (RFE), random forest (RF), and Boruta were used for optimal feature selection. Various classifiers, including artificial neural network (ANN), support vector machine (SVM), and RF, were implemented to create maps of land crops relying on optimal features and training data. After that, in order to increase the result accuracy, maps of land crops derived from several scenarios were fused using a fusion-based voting strategy at the level of decision, and new maps of land crops and classification uncertainty maps were prepared. Subsequently, the performance of different scenarios was evaluated and compared. Among the feature selection algorithms, RF accuracy was higher than RFE and Boruta. Moreover, the efficiency of RF was higher than SVM and ANN. The overall accuracy of the voting scenario was higher than all other scenarios. The finding of this research demonstrated that combining the features’ capabilities extracted from sensors in different spectral ranges, different feature selection algorithms, and classifiers improved the land crop classification accuracy.
•A new method was used for impact assessment of mining activities on environment change.•A remote sensing-based approach was used to predict the LST and NDVI maps.•We applied our method on five mine ...over worldwide.•The forest cover was reduced due to the increasing mining activities.•A significant and negative impact of the mine activities will continue in future.
Mining activities and associated actions cause land-use/land-cover (LULC) changes across the world. The objective of this study were to evaluate the historical impacts of mining activities on surface biophysical characteristics, and for the first time, to predict the future changes in pattern of vegetation cover and land surface temperature (LST). In terms of the utilized data, satellite images of Landsat, and meteorological data of Sungun mine in Iran, Athabasca oil sands in Canada, Singrauli coalfield in India and Hambach mine in Germany, were used over the period of 1989–2019. In the first step, the spectral bands of Landsat images were employed to extract historical LULC changes in the study areas based on the homogeneity distance classification algorithm (HDCA). Thereafter, a CA-Markov model was used to predict the future of LULC changes based on the historical changes. In addition, LST and vegetation cover maps were calculated using the single channel algorithm, and the normalized difference vegetation index (NDVI), respectively. In the second step, the trends of LST and NDVI variations in different LULC change types and over different time periods were investigated. Finally, a CA-Markov model was used to predict the LST and NDVI maps and the trend of their variations in future. The results indicated that the forest and green space cover was reduced from 9.95 in 1989 to 5.9 Km2 in 2019 for Sungun mine, from 42.14 in 1999 to 33.09 Km2 in 2019 for Athabasca oil sands, from 231.46 in 1996 to 263.95 Km2 in 2016 for Singrauli coalfield, and from 180.38 in 1989 to 133.99 Km2 in 2017 for Hambach mine, as a result of expansion and development of of mineral activities. Our findings about Sungun revealed that the areal coverage of forest and green space will decrease to 15% of the total study area by 2039, resulting in reduction of the mean NDVI by almost 0.06 and increase of mean standardized LST from 0.52 in 2019 to 0.61 in 2039. our results further indicate that for Athabasca oil sands (Singrauli coalfield, Hambach mine), the mean values of standardized LST and NDVI will change from 0.5 (0.44 and 0.4) and 0.38 (0.38, 0.35) in 2019 (2016, 2017) to 0.57 (0.5, 0.47) and 0.33 (0.32, 0.28), in 2039 (2036, 2035), respectively. This can be mainly attributed to the increasing mining activities in the past as well as future years. The discussion and conclusions presented in this study can be of interest to local planners, policy makers, and environmentalists in order to observe the damages brought to the environment and the society in a larger picture.
Display omitted
•We proposed a new Land Surface Ecological Status Composition Index (LSESCI)•Biophysical Composition Index (BCI) and Land Surface Temperature (LST) are combined.•LSESCI is compared ...with Remote Sensing-based Ecological Index (RSEI)•LSESCI is superior in modeling SES spatial and temporal variations than RSEI.•SESCI is superior to distinguish SES of different Land Use/Cover (LULC) classes.
Accurate modeling of Land Surface Ecological Status (LSES) is crucial in environmental applications. Despite valuable benefits, common indices are unable to distinguish LSES of bare soils from lands affected by Anthropogenic Destructive Activities (ADAs). The objective of this study was to present an index to distinguish LSES of different Land Use/Covers (LULCs) particularly bare soils from lands affected by ADAs using remote sensing images. Landsat multi-temporal imagery, National Land Cover Database (NLCD), Imperviousness and High Resolution Layer Imperviousness (HRLI) datasets for Arasbaran protected area in Iran and 13 cities from the United States and Europe were used in this study. First, the surface biophysical characteristics and LULC were derived from Landsat images using the single channel algorithm, spectral indices, and support vector machine. Secondly, a new index was developed based on improved Ridd's conceptual Vegetation-Impervious-Soil triangle model and specified as Land Surface Ecological Status Composition Index (LSESCI). LSESCI was developed by combining Biophysical Composition Index (BCI) information and Land Surface Temperature (LST). In the third step, the LSES was modeled based on Remote Sensing-based Ecological Index (RSEI). Variance-based global sensitivity analysis was used to calculate the impact of input parameters on the modeled LSES. Afterwards, the variations in these indices were modeled using Subtraction, Variance and Principal Component Analysis (PCA) strategies. Finally, the efficiency of these indices was assessed and compared to model from the relationships between LSESCI and RSEI with spectral indices, and LULC classes. There was an overall improvement in modelling LSES accuracy using the LSESCI over RSEI. For instance, the difference between the mean RSEI and LSESCI for the lands affected by ADAs and Bare soil lands in Arasbaran protected area in Iran were 0.04 and 0.27, respectively. LST and Wetness have the most and least impact on LSES modeling, respectively, compared to other input parameters. The mean absolute value of the correlation coefficient (r) between greenness, moisture, dryness, and heat indices and LSESCI (RSEI) were 0.90 (0.84), 0.76 (0.69) and 0.93 (0.88), respectively. The mean absolute values of r between variations of different spectral indices and variations of LSESCI (RSEI) obtained from PCA, Variance and Subtraction strategies were 0.89 (0.87), 0.73 (0.64) and 0.79 (0.73), respectively. Similarly, for selected cities in the United States and Europe, the mean of r values between RSEI and LSESCI and NLCD Imperviousness (HRLI) were 0.58 (0.77) and 0.77 (0.85), respectively. Overall, the LSESCI had high ability to distinguish the LSES of different LULC classes especially bare soils from lands affected by ADAs. Thus, the proposed LSESCI was superior in modeling LSES of the urban and non-urban regions with heterogeneous surface over RSEI.
•A new Soil Water Index (SWI) downscaling approach is introduced.•Land surface temperature had the greatest effect on the spatial variation of SWI.•Biophysical properties had greater impact than ...topographic and geographical.•The machine learning based approach showed strong potential in improving SWI.
One of the limitations of daily Soil Water Index (SWI) products obtained from satellite imagery is the low spatial resolution, limiting their precise applications. The purpose of this study was to present a machine learning based approach to improve the spatial resolution of the SWI obtained from the Advanced Scatterometer (ASCAT). Surface biophysical, topographic, and geographical properties (environmental parameters) maps of three field sites from the United States of America (USA), France, and Iran were prepared with a spatial resolution of 30, 1000 and 10,000 m and their effects on SWI were investigated. A SWI estimation model was constructed based on a Random Forest (RF) regression using effective environmental parameters and used to map SWI at 1,000 and 30 m spatial resolutions. The final SWI map with an improved spatial resolution was prepared after applying a correction due to a residual error. Finally, the efficiency of the proposed model was evaluated based on measured soil moisture (SM) data recorded at ground stations. The results showed that land surface temperature had the greatest effect on the spatial distribution of SWI. The impact of surface biophysical properties on the SWI was greater than topographical and geographical properties. The mean SWI error in USA, France, and Iran at spatial resolution of 10,000 (improved 1000 m) for warm season were 23.6 % (15.8 %), 14.2 % (9.8 %) and 10.7 % (7.4 %), respectively. These values for cold season were 27.9 % (17.2 %), 15.3 % (13.2 %) and 15.5 % (8.8 %), respectively. Mean of R2 and RMSE between measured SM values and SWI 10,000 m (1000 m and 30 m) were 0.13 (0.43 and 0.73) and 17.6 (12.1 and 7.2 %), respectively. These values for cold season were 0.10 (0.39, 0.67), and 20.7 (14.3, 7.2 %), respectively. The proposed machine learning based approach showed strong potential in improving the spatial resolution of SWI and giving the opportunity for various precise applications.
Urban land surface temperature (ULST) is a key variable for environmental applications and widely used in a variety of fields. However, retrieving ULST with high spatial and temporal resolution is ...still a challenging task since its a function of surface characteristics and geographical, climatic, and seasonal conditions. Assessing the type and intensity of the impact of surface characteristics effect on ULST is one of great importance way in improving urban environmental conditions. Hence, the purpose of this study is to evaluate and model the impact of surface characteristics and their adjacency effects on ULST in different seasonal conditions and latitudes. First, ULST and surface characteristic maps were prepared for six European cities in different seasonal conditions. Then, adjacency effects maps of these surface characteristicswere prepared using the proposed inverse distance weighted kernel. Finally, the impacts of surface characteristics and their adjacency effects on ULST were evaluated and compared in different seasonal conditions and latitudes. The mean absolute r between biophysical (topographical) characteristics and ULST in warm and cold seasons were 0.52 (0.22) and 0.25 (0.12), respectively. In the warm season, the mean increase in absolute r between the surface biophysical and topographical characteristics and ULST due to the adjacency effects were 0.10 and 0.04, respectively. For the cold season, these values were 0.03 and 0.03, respectively. By increasing latitude, the impact of surface biophysical characteristics on ULST increased, and the impact of topographical characteristics decreased. ULST modeling accuracy increases by consideration of the adjacency effects. ULST was affected more by the adjacency effects of surface biophysical characteristics than the adjacency effects of surface topographical characteristics. The results of this study showed that in evaluating the impact of surface characteristics on ULST, their adjacency effects and seasonal and geographical conditions should be taken into account.
Display omitted
•An inverse distance weighted kernel was formulated to quantify the effective surface characteristics (adjacency effects).•ULST presented spatial and seasonal variations across six cities of European capitals in different latitudes.•ULST is significantly affected by environmental variables and their adjacency effects.•ULST in the warm (cold) season seem to be more affected by biophysical (topographical) variables than the cold (warm) season.•By increasing latitude, biophysical variables impact on ULST increased, and topographical variables impact decreased.
One of the key parameters that affects the accuracy of land surface temperature (LST) disaggregation is the environmental variables that are fed to the disaggregation model. The aim of this article ...is to present a new strategy for the disaggregation of LST based on adjacency effects. To do this, a dataset obtained from satellite images and auxiliary information from five European cities was used. First, maps of environmental variables that affect LST were collected. Second, a map of effective environmental variables was produced by calculating and applying the influence of the adjacency effects of each environmental variable based on the proposed weighted inverse distance kernel. Finally, the datasets of environmental variables and effective environmental variables were used separately in the disaggregation process to convert LST at 990 m to disaggregated LST (DLST) at 90 m. The mean RMSEs between LST and DLST obtained without considering the adjacency effects approach for the built-up, agricultural, pasture, forest, and water lands in the cold (warm) season were 0.85 (1.55), 0.72 (1.31), 0.98 (1.63), 0.59 (1.2), and 0.40 (1.12) K, respectively. Taking into account the adjacency effects, the mean RMSE between LST and DLST on built-up, agricultural, pasture, forest, and water lands used in the cold season decreased by 0.35, 0.17, 0.13, 0.09, and 0.03 K, respectively. These values were 0.54, 0.36, 0.33, 0.34, and 0.07 K for the warm season, respectively. The result showed that considering adjacency effects increases the accuracy of LST disaggregation.
Urban Surface Ecological Status (USES) reflects the structure and function of an urban ecosystem. USES is influenced by the surface biophysical, biochemical, and biological properties. The assessment ...and modeling of USES is crucial for sustainability assessment in support of achieving sustainable development goals such as sustainable cities and communities. The objective of this study is to present a new analytical framework for assessing the USES. This analytical framework is centered on a new index, Remotely Sensed Urban Surface Ecological index (RSUSEI). In this study, RSUSEI is used to assess the USES of six selected cities in the U.S.A. To this end, Landsat 8 images, water vapor products, and the National Land Cover Database (NLCD) land cover and imperviousness datasets are downloaded for use. Firstly, Land Surface Temperature (LST), Wetness, Normalized Difference Vegetation Index (NDVI), and Normalized Difference Soil Index (NDSI) are derived by remote sensing methods. Then, RSUSEI is developed by the combination of NDVI, NDSI, Wetness, LST, and Impervious Surface Cover (ISC) with Principal Components Analysis (PCA). Next, the spatial variations of USES across the cities are evaluated and compared. Finally, the association degree of each parameter in the USES modeling is investigated. Results show that the spatial variability of LST, ISC, NDVI, NDSI, and Wetness is heterogeneous within and between cities. The mean (standard deviation) value of RSUSEI for Minneapolis, Dallas, Phoenix, Los Angeles, Chicago and Seattle yielded 0.58 (0.16), 0.54 (0.17), 0.47 (0.19), 0.63 (0.21), 0.50 (0.17), and 0.44 (0.19), respectively. For all the cities, PC1 included more than 93% of the surface information, which is contributed by greenness, moisture, dryness, heat, and imperviousness. The highest and lowest mean values of RSUSEI are found in “Developed, High intensity” (0.76) and “Developed, Open Space” (0.35) lands, respectively. The mean correlation coefficient between RSUSEI and LST, ISC, NDVI, NDSI, and Wetness, is 0.47, 0.97, −0.31, 0.17, and −0.27, respectively. The statistical significance of these correlations is confirmed at 95% confidence level. These results suggest that the association degree of ISC in USES modeling is the highest, despite the differences in land cover and biophysical characteristics in the cities. RSUSEI could be very useful in modeling and comparing USES across cities with different geographical, climatic, environmental, and biophysical conditions and can also be used for assessing urban sustainability over space and time.
The increase of Land Surface Temperature (LST) and the formation of heat island in megacities have become an emerging environmental concern. The main objective of this study is to predict the ...intensity of Tehran's heat island in the year 2033 based on historical changes of land cover and LST. For this purpose, Landsat satellite images were integrated with meteorological stations' measurements from 1985 to 2017. The Cellular Automata-Markov (CA-Markov) and Artificial Neural Network (ANN) models were used to predict the land cover changes and to the modelling of the Surface Urban Heat Island Intensity (SUHII), Surface Urban Heat Island Ratio Index (SUHRI) was used. Subsequently, using statistical analysis of the effect of historical land cover changes on LST variations, SUHII for 2033 was predicted. Our findings show that within this period, the built-up lands increased significantly from 39% in 1985 to 65% in 2017. The intensity of heat island increased with an increase in the value of SUHII from 0.02 to 0.19. Our predictive analysis reveals that the intensity of the Tehran's heat island will increase to 0.32 by 2033. Our conclusions draw attentions to the increasing LST now and in the future in Tehran so that urban planners and local authorities take adequate actions for controlling its environmental impacts.
The Tropical Rainfall Measuring Mission (TRMM) and Global Precipitation Mission (GPM) are the most important and widely used data sources in several applications—e.g., forecasting drought and flood, ...and managing water resources—especially in the areas with sparse or no other robust sources. This study explored the accuracy and precision of satellite data products over a span of 18 years (2000–2017) using synoptic ground station data for three regions in Iran with different climates, namely (a) humid and high rainfall, (b) semi-arid, and (c) arid. The results show that the monthly precipitation products of GPM and TRMM overestimate the rainfall. On average, they overestimated the precipitation amount by 11% in humid, by 50% in semi-arid, and by 43% in arid climate conditions compared to the ground-based data. This study also evaluated the satellite data accuracy in drought and wet conditions based on the standardized precipitation index (SPI) and different seasons. The results showed that the accuracy of satellite data varies significantly under drought, wet, and normal conditions and different timescales, being lowest under drought conditions, especially in arid regions. The highest accuracy was obtained on the 12-month timescale and the lowest on the 3-month timescale. Although the accuracy of the data is dependent on the season, the seasonal effects depend on climatic conditions.
Due to the excessive use of natural resources in the contemporary world, the importance of ecological and environmental condition modeling has increased. Wetlands and cities represent the natural and ...artificial strategic areas that affect ecosystem conditions. Changes in the ecological conditions of these areas have a great impact on the conditions of the global ecosystem. Therefore, modeling spatiotemporal variations of the ecological conditions in these areas is critical. This study was aimed at comparing degrees of variation among surface ecological conditions due to natural and unnatural factors. Consequently, the surface ecological conditions of Gomishan city and Gomishan wetland in Iran were modeled for a period of 30 years, and the spatiotemporal variations were evaluated and compared with each other. To this end, 20 Landsat 5, 7, and 8, and 432 Moderate Resolution Imaging Spectroradiometer (MODIS), monthly land surface temperature (LST) (MOD11C3) and normalized difference vegetation index (NDVI) (MOD13C3) products were utilized. The surface ecological conditions were modeled according to the Remote Sensing-based Ecological Index (RSEI), and the spatiotemporal variation of the RSEI values in the study area (Gomishan city, Gomishan wetland) were evaluated and compared with each other. According to MODIS products, the mean of the LST and NDVI variance values for the study area (Gomishan city, Gomishan wetland) were obtained to be 6.5 °C (2.1, 12.1) and 0.009 (0.005, 0.013), respectively. The highest LST and NDVI temporal variations were found for Gomishan wetland near the Caspian Sea. According to Landsat images, Gomishan wetland and Gomishan city have the highest and lowest temporal variations in surface biophysical characteristics, respectively. The mean RSEI for the study area (Gomishan city, Gomishan wetland) was 0.43 (0.65, 0.29), respectively. Additionally, the mean Coefficient of Variation (CV) of RSEI for the study area (Gomishan city, Gomishan wetland) was 0.10 (0.88, 0.51), respectively. The surface ecological conditions of Gomishan city were worse than those of the Gomishan wetland at all dates. Temporal variations in the surface ecological conditions of Gomishan wetland were greater than those of the study area and Gomishan city. These results can provide useful and effective information for environmental planning and decision-making to improve ecological conditions, protect the environment, and support sustainable ecosystem development.