Spatio-temporal Dynamics of the Standardized Precipitation Index (SPI) and the Normalized Difference Vegetation Index (NDVI) for the Characterization of Drought in Côte d’Ivoire

2024-07-19T11:22:15+00:00

This study aims to use the potential of satellite imagery for monitoring and characterization of drought conditions based on the Standardized Precipitation Index (SPI) and the Normalized Difference Vegetation Index (NDVI) in the Côte d'Ivoire from 2010-2015 during the season from November to March. These indexes were obtained from two types of data, namely rainfall data from the Tropical Rainfall Measuring Mission (TRMM) and NDVI data from the Moderate Resolution Imaging Spectroradiometer (MODIS) sensor on board the Earth Observation System (EOS) Terra satellite and the EOS-Aqua platform. They were combined to perform a spatio-temporal analysis of the Drought Index (DI) to determine the driest months. Analyzes and digital processing of the data using the ArgGis 10.2.1 software showed that the months of December and January are the most critical months according to the SPI analysis. On the other hand, according to the NDVI, the critical months are January and February. As for the integrated Drought Index (DI), it indicates that the moderately humid months observed in November, December and March are respectively the mark of the end and the beginning of the rainy season in Côte d'Ivoire. Clearly, the months of January and February are the least rained and therefore the driest and the month most affected by drought is that of January (-0.46 ≤ DI ≤ -0.18) in Côte d'Ivoire during the study period.

Keywords DI; Drought; Ivory Coast; NDVI; SPI

Enhancing Semantic Segmentation of Cloud Images Captured with Horizon-Oriented Cameras

2024-06-03T11:08:06+00:00

The segmentation of sky cloud images is a complex task essential for applications like weather analysis. Compared to all-sky imagers, horizon-oriented cameras provide a more detailed view of clouds near the horizon. In our study, we evaluated three semantic segmentation models: HRNet48, PPLite, and SegFormerB3, utilizing a variety of loss functions on a novel dataset of horizon cloud images. Throughout our experiments, we consistently observed segmentation leakage issues. To address this, we introduced machine learning-based post-processing methods, including random forest and xgboost, that leverage region-specific features to refine the segmentation. Our results showed notable improvements, with the Cumuliform class dice score increasing from 0.552 to 0.583, and Stratiform class accuracy improving from 0.49 to 0.511 when applying xgboost on SegFormerB3's output. The study revealed the relative contributions of the loss functions and post-processing steps.

Keywords Remote Sensing; Segmentation; Sky Clouds; Deep Learning

International Journal of Advanced Remote Sensing and GIS

Spatio-temporal Dynamics of the Standardized Precipitation Index (SPI) and the Normalized Difference Vegetation Index (NDVI) for the Characterization of Drought in Côte d’Ivoire

Enhancing Semantic Segmentation of Cloud Images Captured with Horizon-Oriented Cameras