Focal-TSMP: deep learning for vegetation health prediction and agricultural drought assessment from a regional climate simulation

Geoscientific model development, Journal Year: 2024, Volume and Issue: 17(7), P. 2987 - 3023

Published: April 16, 2024

Abstract. Satellite-derived agricultural drought indices can provide a complementary perspective of terrestrial vegetation trends. In addition, their integration for assessments under future climates is beneficial providing more comprehensive assessments. However, satellite-derived are only available the Earth observation era. this study, we aim to improve climate change by applying deep learning (DL) predict from regional simulation. The simulation produced Terrestrial Systems Modeling Platform (TSMP) and performed in free evolution mode over Europe. TSMP simulations incorporate variables underground top atmosphere (ground-to-atmosphere; G2A) widely used research studies related water cycle change. We leverage these long-term forecasting DL map forecast into normalized difference index (NDVI) brightness temperature (BT) images that not part model. These predicted then derive different indices, namely NDVI anomaly, BT condition (VCI), thermal (TCI), health (VHI). developed model could be integrated with data assimilation downstream tasks, i.e., estimating periods where no satellite modeling impact extreme events on responses scenarios. Moreover, our study as evaluation framework TSMP-based simulations. To ensure reliability assess model’s applicability seasons regions, an analysis biases uncertainties across regions pan-European domain. further about contribution input components better understanding prediction. A using reference remote sensing showed sufficiently good agreements between predictions observations. While performance varies test set it achieves mean absolute error (MAE) 0.027 1.90 K coefficient determination (R2) scores 0.88 0.92 BT, respectively, at 0.11° resolution sub-seasonal predictions. summary, demonstrate feasibility synthesize images, which forecasting. Our implementation publicly project page (https://hakamshams.github.io/Focal-TSMP, last access: 4 April 2024).

Language: Английский

---

First Analyses of the TIMELINE AVHRR SST Product: Long-Term Trends of Sea Surface Temperature at 1 km Resolution across European Coastal Zones

Remote Sensing, Journal Year: 2024, Volume and Issue: 16(11), P. 1932 - 1932

Published: May 27, 2024

Coastal areas are among the most productive in world, ecologically as well economically. Sea Surface Temperature (SST) has evolved major essential climate variable (ECV) and ocean (EOV) to monitor land–ocean interactions oceanic warming trends. SST monitoring can be achieved by means of remote sensing. The current relatively coarse spatial resolution established products limits their potential small-scale, coastal zones. This study presents first analysis TIMELINE 1 km product from AVHRR four key European regions: Northern Baltic Sea, Adriatic Aegean Balearic Sea. monthly anomaly trends showed high positive all areas, exceeding global average warming. Seasonal variations reveal peak during spring, early summer, autumn, suggesting a seasonal shift. revealed significantly higher at near-coast which were especially distinct Mediterranean areas. clearest pattern was visible March May, where coast twice that observed 40 distance coast. To validate our findings, we compared time series with anomalies derived Level 4 CCI product. comparison an overall good accordance correlation coefficients R > 0.82 for = 0.77 North Seas. highlights Local Area Coverage (LAC) data mapping long-term variability, such regions.

Language: Английский

---

Seasonal Vegetation Trends in Biomes of Türkiye: A Decade-Long Analysis Using NDVI Time Series

Bartın Orman Fakültesi Dergisi, Journal Year: 2024, Volume and Issue: unknown

Published: July 16, 2024

This study analyzes Türkiye's biomes' seasonal vegetation trend from 2014 to 2023 using the Normalized Difference Vegetation Index (NDVI) and Google Earth Engine (GEE). Focusing on Mediterranean Forests, Woodlands & Scrub; Temperate Broadleaf Mixed Forests; Grasslands, Savannas Shrublands; Coniferous Forests biomes, it aims illuminate vegetative trends inform conservation strategies in line with European Green Deal. Using Landsat 8 Operational Land Imager (OLI) satellite imagery GEE's computational capabilities, efficiently processes large datasets, revealing distinctive responses climatic conditions across biomes. Key findings include resilience of drought, stable growth temperate broadleaf mixed forests, dynamic shifts grasslands, consistent photosynthetic activity coniferous forests. The highlights importance continuous monitoring suggests future research integrating remote sensing ground observations for ecosystem management under climate change.

Language: Английский

---

Surface water dynamics of Lake Chad Basin (Sahelian Africa) based on daily temporal resolution earth observation time series

Journal of Hydroinformatics, Journal Year: 2024, Volume and Issue: 26(9), P. 2325 - 2352

Published: Aug. 29, 2024

ABSTRACT Water availability is vital for the sustenance of livelihoods in Lake Chad Basin. However, daily and seasonal dynamics open water bodies are not well understood. This study aims to (1) analyze bodies, (2) estimate changes surface area extent including trends change points, (3) assess connection between rainfall variation. To achieve this, we used Global WaterPack ERA5-Land aggregated datasets. We employed time series decomposition, analysis, temporal lag correlation our analysis. The results showed strong patterns natural lakes compared reservoirs/dams. Between 2003 2022, averaged 2,475.64 km2. Northern pool exhibited significant fluctuations, remaining below 600 km² 2005 2012, from 2016 2019), with less than 350 km2 lasting only a few days annually. Southern 2,200 2,400 km2, except during drought years (2006–2007), specifically year approximately 66, 301–365/6. In Fitri, yearly maximum minimum extents were observed 1–59 305–365/6, 60 304, respectively.

Language: Английский

---

Application of Google Earth Engine NDVI Trend to Study Yield of Sugarcane Crop Using Sentinel 2 Data

Advances in social networking and online communities book series, Journal Year: 2024, Volume and Issue: unknown, P. 309 - 342

Published: Nov. 1, 2024

Increasing global food demand and the effects of climate change further indicate a need for proper ways monitoring crop health. This chapter has demonstrated importance normalized difference vegetation index (NDVI) as non-invasive means A review literature indicates that NDVI is useful in determining stress, diseases, performance, especially if considered on long-term basis. study based sugarcane Vuyyuru Village, Andhra Pradesh, considering to analyze health five-year period 2018-2022. In this chapter, pre-processing Sentinel satellite imagery through atmospheric correction image registration was carried out ensure data accuracy ensured. The computation values each year involves assessing any patterns or variations are found spatially. work sets enhance understanding dynamics time, thus giving valued insights future agricultural management.

Language: Английский

---