Negative impacts of climate change on ecosystems have been increasing, and both the intensification and the mitigation of these impacts are strongly linked with human activities. Management and reduction of human-induced disturbances on ecosystems can mitigate the effects of climate change and enhance the ecosystem recovery process. Here, we investigate coupled human and climate effects on the wetland ecosystem of the lower Helmand basin from 1977 to 2014. Using time series climate-variable data and land-use changes from Landsat time series imagery, we compared changes in ecosystem status between the upstream and downstream regions. Results show that despite a strong and prolonged drought in the region, the upstream region of the lower Helmand basin remained dominated by agriculture, causing severe water stress on the Hamoun wetlands downstream. The loss of available water in wetlands was followed by large-scale land abandonment in rural areas, migration to the cities, and increasing unemployment and economic hardship. Our results suggest that unsustainable land-use policies in the upstream region, combined with synergistic effects of human activities and climate in lower Helmand basin, have exacerbated the effects of water stress on local inhabitants in the downstream region.

Accurate crop type mapping using satellite imagery is crucial for food security, yet accurately distinguishing between crops with similar spectral signatures is challenging. This study assessed the performance of Sentinel-2 (S2) time series (spectral bands and vegetation indices), Sentinel-1 (S1) time series (backscattering coefficients and polarimetric parameters), alongside phenological features derived from both S1 and S2 time series (harmonic coefficients and median features), for classifying sunflower, soybean, and maize. Random Forest (RF), Multi-Layer Perceptron (MLP), and XGBoost classifiers were applied across various dataset configurations and train-test splits over two study sites and years in France. Additionally, the InceptionTime classifier, specifically designed for time series data, was tested exclusively with time series datasets to compare its performance against the three general machine learning algorithms (RF, XGBoost, and MLP). The results showed that XGBoost outperformed RF and MLP in classifying the three crops. The optimal dataset for mapping all three crops combined S1 backscattering coefficients with S2 vegetation indices, with comparable results between phenological features and time series data (mean F1 scores of 89.9% for sunflower, 76.6% for soybean, and 91.1% for maize). However, when using individual satellite sensors, S1 phenological features and time series outperformed S2 for sunflower, while S2 was superior for soybean and maize. Both phenological features and time series data produced close mean F1 scores across spatial, temporal, and spatiotemporal transfer scenarios, though median features dataset was the best choice for spatiotemporal transfer. Polarimetric S1 data did not yield effective results. The InceptionTime classifier further improved classification accuracy over XGBoost for all crops, with the degree of improvement varying by crop and dataset (the highest mean F1 scores of 90.6% for sunflower, 86.0% for soybean, and 93.5% for maize).

The complex, dynamic and narrow boundaries between vegetation types make wetland mapping challenging. Hereafter the case study of the Hamoun-e-Hirmand wetland is considered by analysing eight Synthetic Aperture Radar (SAR) Images acquired in dry and wet periods with three wavelengths (X-band ~ 3 cm, C-band ~ 6 cm, and L-band ~ 25 cm), three polarizations (HH, VV and VH), and four incidence angles (22°, 30°, 34° and 53°). Then, the Support Vector Machine (SVM) classification method was applied to classify TerraSAR-X, Sentinel-1, and ALOS-PALSAR images. The final wetland land cover map was created by combining the classification results obtained from each sensor. In the case in question, results show that TerraSAR-X (X-band, HH-53°) and Sentinel-1 data (C-band, VV-34°) were useful for determining the flooded vegetation area in the wet period. This is crucial for the conservation of water bird habitats since flooded vegetation is an ideal environment for the nesting and feeding of water birds. PALSAR data (L-band in both HH and VH polarizations, 30°) were capable of separating the classes of vegetation density in the wetland. In the dry period, Sentinel-1 (VV and VH, 34°) and TerraSAR-X (HH, 22° and 53°) had higher potential in land cover mapping than PALSAR (HH and VH, 30°). Based on these results, Sentinel-1 in VV and VH provides the highest ability to discriminate between dry and green plants. TerraSAR-X is better for separating meadow and bare land. The results obtained in this paper can reduce the ambiguity in selecting satellite data for wetland studies. The results can also be used to produce more accurate data from satellite images and to facilitate wetland investigation, conservation and restoration.