Estimating surface NO2 concentrations over Europe using Sentinel-5P TROPOMI observations and Machine Learning

Remote Sensing of Environment, Journal Year: 2024, Volume and Issue: 312, P. 114321 - 114321

Published: July 24, 2024

Satellite observations from instruments such as the TROPOspheric Monitoring Instrument (TROPOMI) show significant potential for monitoring spatiotemporal variability of NO2, however they typically provide vertically integrated measurements over tropospheric column. In this study, we introduce a machine learning approach entitled 'S-MESH' (Satellite and ML-based Estimation Surface air quality at High resolution) that allows estimating daily surface NO2 concentrations Europe 1 km spatial resolution based on eXtreme gradient boost (XGBoost) model using primarily observation-based datasets period 2019–2021. Spatiotemporal used by include TROPOMI vertical column density, night light radiance Visible Infrared Imaging Radiometer Suite (VIIRS), Normalized Difference Vegetation Index Moderate Resolution Spectroradiometer (MODIS), stations European Environment Agency database modeled meteorological parameters planetary boundary layer height, wind velocity, temperature. The overall evaluation shows mean absolute error 7.77 μg/m3, median bias 0.6 μg/m3 Spearman rank correlation 0.66. performance is found to be influenced concentration levels, with most reliable predictions levels 10–40 <40%. temporal analyses indicate robustness across study area, better prediction accuracy during winter months associated higher concentrations. Despite complexity continental scale XGBoost-based fast execution in providing estimates Europe. Shapley Additive exPlanations (SHAP) value analysis highlights density main source information deriving concentrations, indicating its studies. SHAP values also importance anthropogenic emission proxy inputs VIIRS lights, complementing detailed patterns variations.

Language: Английский

---

Monitoring Agricultural Land Loss by Analyzing Changes in Land Use and Land Cover

Emerging Science Journal, Journal Year: 2024, Volume and Issue: 8(2), P. 687 - 699

Published: April 1, 2024

The agricultural sector's output holds paramount significance for the global population, serving as an indispensable resource survival and consumption. Consequently, alterations in landscapes bear substantial implications world's food supply. objectives of this research are to investigate depletion land, with a specific focus on Samut Songkhram Province—an agriculturally prominent region Thailand renowned supplying seafood fruits Bangkok. By employing advanced remote sensing change detection methods incorporating indices like NDVI, NDWI, NDBI, study meticulously analyzed land-use changes. outcomes were rigorously scrutinized through supervised classification, validated by on-site inspections, corroborated data from pertinent agencies. Findings revealed that had sustained its prominence constituting around 70% province's total area over past two decades. However, expanse has undergone persistent transformation during last 20 years. Notably, most surge was observed conversion land urban developed areas, particularly zones Amphawa District, followed Mueang Bang Khonthi districts. This investigation illuminates consistent downward trend vital source sustenance Thailand's population community. Doi: 10.28991/ESJ-2024-08-02-020 Full Text: PDF

Language: Английский

---

Vegetation cover change in Ugam Chatkal National Park, Uzbekistan, in relation to climate variables during the post-Soviet period (1991-2022)

Forum Geografi, Journal Year: 2024, Volume and Issue: 38(1)

Published: March 16, 2024

This paper presents a comprehensive study relating to the vegetation cover change in Ugam Chatkal National Park (Uzbekistan) and its relation climate during post-Soviet period (1991-2022). The utilises remote sensing technology, specifically Normalised Difference Vegetation Index (NDVI) Soil-Adjusted (SAVI), monitor spatio-temporal changes vegetation. Landsat satellite imagery meteorological data, including temperature precipitation records form basis of analysis. research aims understand impact climatic factors, such as air temperature, soil on cover. Statistical methods, for example Pearson’s correlation analysis are employed determine strength direction relationships between these variables. reveals that both NDVI SAVI strongly correlated with temperatures, indicating significant influence factors health growth. findings suggest closely tied change, revealing substantial time, trend towards increasing temperatures. also forecasts future trends, predicting an increase over next four decades. In particular, highlights magnitude monitoring understanding complex interactions dynamics, which crucial environmental management policy-making region.

Language: Английский

---

Urban Ecological Quality Assessment Based on Google Earth Engine and Driving Factors Analysis: A Case Study of Wuhan City, China

Sustainability, Journal Year: 2024, Volume and Issue: 16(9), P. 3598 - 3598

Published: April 25, 2024

Ecological quality is a critical factor affecting the livability of urban areas. Remote sensing technology enables rapid assessment ecological (EQ), providing scientific theoretical support for maintenance and management ecology. This paper evaluates analyzes EQ its driving factors in city Wuhan using remote data from five periods: 2001, 2006, 2011, 2016, 2021, supported by Google Earth Engine (GEE) platform. By employing principal component analysis, Sensing Index (RSEI) was constructed to assess spatiotemporal differences City. Furthermore, study utilized optimal parameter-based geographical detector model analyze influence such as elevation, slope, aspect, population density, greenness, wetness, dryness, heat on RSEI value 2021 further explored impact changes precipitation temperature Wuhan. The results indicate that (1) analysis shows greenness wetness positively affect Wuhan’s EQ, while dryness have negative impacts; (2) reveals 2001 showed trend initial decline followed improvement, with classification grades evolving poor average good better; (3) all nine indicators certain Wuhan, ranking NDVI > NDBSI LST WET elevation density GDP slope aspect; (4) annual non-significant EQ. has improved recent years, but comprehensive still requires enhancement.

Language: Английский

---

Correction of Sun-View Angle Effect on Normalized Difference Vegetation Index (NDVI) with Single View-Angle Observation

IEEE Transactions on Geoscience and Remote Sensing, Journal Year: 2024, Volume and Issue: 62, P. 1 - 13

Published: Jan. 1, 2024

Language: Английский

---

Land use and landscape pattern changes in the Fenhe River Basin, China

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: Jan. 20, 2025

The composition and pattern of ecosystems play a crucial role in determining the overall condition spatial variations ecosystem services. In this study, we explored Normalized Difference Vegetation Index (NDVI), six land use/land cover change (LULC) types, their landscape patterns to reflect spatial-temporal dynamics from 2010 2020 upper middle reaches Fenhe River Basin. trend analysis Mann-Kendall tests was used assess NDVI variation each pixel over past decade. Shannon's Diversity (SHDI) quantify complexity local LULC. Meanwhile, dynamic vegetation LULC types were illustrated using geographic mapping methods. results presented as follows: (1) From 2020, increased, with an annual increment 0.003 (P < 0.001). Spatially, there significant increasing NDVI, particularly high-altitude mountainous areas where mostly consisted forests or grasslands. However, no low-lying urban agglomeration area. (2) During period decrease area water bodies wetlands, rate slowed 2015 2020. Specifically, percentage increased by more than 14%. This proportion exceeded 30% when compared final governmental ecological management goal set during 13th Five-Year Plan Period (2016–2020). (3) same period, SHDI exhibited heterogeneous characteristics. near river basin boundary, values below 1.0. Conversely, low-altitude plains frequent human activities, 1.0, reaching above 1.5 areas. study provides scientific reference for construction environment Basin, along practice insights protection high-quality development Yellow

Language: Английский

---

Lodgepole pine (Pinus contorta Douglas ex Loudon) invasion in subarctic Iceland: evidence from a long-term study

NeoBiota, Journal Year: 2025, Volume and Issue: 97, P. 47 - 66

Published: Jan. 27, 2025

The North American lodgepole pine ( Pinus contorta ) has been widely introduced globally and is now considered invasive in several countries. It was first planted subarctic Iceland the 1950s. Recently, forestry sector strongly promoted it as an attractive means of carbon capture to mitigate global climate change. most extensively tree species Iceland. We describe expansion from a mid-20 th -century plantation Steinadalur, southeast Iceland, decadal changes between 2010 2021. extent occurrence expanded nearly tenfold, with number population density reflecting exponential growth patterns. colonised diverse habitats, including native birch woodlands heathland, associated significant reductions vascular plant richness diversity. conclude that characteristics Steinadalur this will also apply many ecosystems across lowland regions Our study highlights urgent need for management strategies long-term ecological impacts invasion environments.

Language: Английский

---

Integration of remote sensing and machine learning algorithm for agricultural drought early warning over Genale Dawa river basin, Ethiopia

Environmental Monitoring and Assessment, Journal Year: 2025, Volume and Issue: 197(3)

Published: Feb. 4, 2025

Language: Английский

---

Evaluation of Scale Effects on UAV-Based Hyperspectral Imaging for Remote Sensing of Vegetation

Remote Sensing, Journal Year: 2025, Volume and Issue: 17(6), P. 1080 - 1080

Published: March 19, 2025

With the rapid advancement of unmanned aerial vehicles (UAVs) in recent years, UAV-based remote sensing has emerged as a highly efficient and practical tool for environmental monitoring. In vegetation sensing, UAVs equipped with hyperspectral sensors can capture detailed spectral information, enabling precise monitoring plant health retrieval physiological biochemical parameters. A critical aspect is accurate acquisition canopy reflectance. However, due to mobility variation flight altitude, data are susceptible scale effects, where changes spatial resolution significantly impact This study investigates issue UAV imaging, focusing on how varying altitudes influence atmospheric correction, viewer geometry, heterogeneity. Using images captured at different Chinese fir forest stand, we propose two correction methods: one based uniform grey reference panel same altitude another altitude-specific panels. The reflectance spectra indices, including NDVI, EVI, PRI, CIRE, were computed analyzed across altitudes. results show significant variations indices lower altitudes, NDVI CIRE demonstrating largest between 50 m 100 m, heterogeneous structure near-infrared scattering. For instance, increased by 18% from 75 stabilized after while standard deviation decreased 32% 250 indicating reduced heterogeneity effects. Similarly, PRI exhibited notable increases attributed shadowing soil background proportions, stabilizing above m. Above diminished, became minimal (<3%), although geometry effects persisted. These findings emphasize that conducting observations least minimizes ensuring more consistent reliable highlights importance standardized protocols optimal selection improve accuracy comparability data, contributing advancements carbon estimation.

Language: Английский

---

Effect of natural and anthropogenic factors on the vegetation distribution in central Loess Plateau, China

Journal of Mountain Science, Journal Year: 2025, Volume and Issue: unknown

Published: March 26, 2025

Language: Английский

---