Site environments and instrumental characteristics of carbon dioxide (CO₂) measurements operated by local governments in the Kanto Plain, the center of which is Tokyo were summarized for this study. The observation sites were classified into environments of three types: urban, suburban, and woodland. Based on a few decades of accumulated hourly data, the diurnal and seasonal variations of CO₂ concentrations were analyzed as a composite of anomalies from annual means recorded for each site. In urban areas, the highest concentrations appear before midnight in winter. The second peak corresponds to the morning rush hour and strengthening of the inversion layer. Suburban areas can be characterized as having the highest concentration before dawn and the lowest concentration during the daytime in summer in association with the activation of respiration and photosynthesis of vegetation. In these areas, concentration peaks also appear during the morning rush hour. Woodland areas show background features, with the highest concentration in early spring: higher than the global background by about 5 ppmv.

This study investigates the capabilities of high-resolution Sentinel-2 observations and deep learning (DL) segmentation models, specifically convolutional neural networks (CNNs), for accurate mapping of small and fragmented agricultural burn areas in Punjab, India. Initially, the model was trained using ICNF burn area data from Portugal to capture large fire and burn area delineation, achieving moderate accuracy. Subsequent fine-tuning using annotated data from Punjab improved the model’s ability to detect small burn patches, demonstrating higher accuracy and precision than the baseline Normalised Burn Ratio (NBR) Index method. On-ground validation using buffer zone analysis and crop field images confirms the DL approach effectiveness. Cloud interference and temporal gaps in satellite data posed some challenges. Despite these limitations, the study highlights the methodological advancements and potential of using DL models for small burn area detection in agricultural settings. The model achieved overall testing accuracy of 0.64 in Dice score assessment. This report presents a promising approach for improved mapping of the burn area, which might engender better estimation of emissions and the management practices in regions with little available ground truth data.