Remote sensing and AI-based deforestation and reforestation potential analysis in the wake of climate change - A case study of flood affected regions

Abstract

This research investigates the capabilities of remote sensing and AI-based analysis in the context of deforestation and reforestation within flood-affected regions, highlighting the impact of climate change on forest cover. The study utilized Sentinel-2 data to collect in formation for July-August across 2018-2022, and May-June and November-December of 2022 for pre- and post-flood analysis, respectively. Due to a lack of annotated data, the study employed unsupervised learning techniques, including Kmeans++, and deep em bedded clustering, to analyze the data. Various features such as Normalized Difference Vegetation Index (NDVI), Soil-Adjusted Vegetation Index (SAVI), Shadow Index (SI), Advanced Vegetation Index (AVI), and Normalized Difference Water Index (NDWI) were employed to identify deforested areas in Pakistan due to floods. The results of the study revealed that Deep Embedded Clustering (DEC) outperformed traditional machine learning methods like Kmeans++ and Kmeans, indicating its po tential for future research in this field. The November 2022 - December 2022 data was also used to determine the potential areas of reforestation. The findings of this study demonstrate the potential of remote sensing and AI-based analysis for deforestation and reforestation in flood-affected regions. The results obtained indicate a reduction in tree cover over the years, consistent with official figures reported by the UN and the Pakistan Forest Institute. Additionally, our findings reveal fluctuations in forest cover before and after the monsoon season in 2022, highlighting the importance of monitoring and conservation efforts in flood-affected regions. The study provides valuable insights for policymakers and environmentalists in the fight against climate change, highlighting the importance of preserving forest cover and pro moting reforestation efforts. This research can serve as a foundation for future studies in this field, paving the way for innovative solutions that address the challenges posed by climate change.