Human degradation of tropical moist forests is greater than previously estimated

Nature, Journal Year: 2024, Volume and Issue: 631(8021), P. 570 - 576

Published: July 3, 2024

Abstract Tropical forest degradation from selective logging, fire and edge effects is a major driver of carbon biodiversity loss 1–3 , with annual rates comparable to those deforestation 4 . However, its actual extent long-term impacts remain uncertain at global tropical scale 5 Here we quantify the magnitude persistence multiple types on structure by combining satellite remote sensing data pantropical moist cover changes estimates canopy height biomass spaceborne 6 light detection ranging (LiDAR). We estimate that decreases owing logging 15% 50%, respectively, low recovery even after 20 years. Agriculture road expansion trigger 20% 30% reduction in edge, persistent being measurable up 1.5 km inside forest. Edge encroach 18% (approximately 206 Mha) remaining forests, an area more than 200% larger previously estimated 7 Finally, degraded forests 50% are significantly vulnerable subsequent deforestation. Collectively, our findings call for greater efforts prevent protect already meet conservation pledges made recent United Nations Climate Change Biodiversity conferences.

Language: Английский

---

A time-continuous land surface temperature (LST) data fusion approach based on deep learning with microwave remote sensing and high-density ground truth observations

The Science of The Total Environment, Journal Year: 2024, Volume and Issue: 914, P. 169992 - 169992

Published: Jan. 10, 2024

Language: Английский

---

Satellite Data in Agricultural and Environmental Economics: Theory and Practice

Agricultural Economics, Journal Year: 2025, Volume and Issue: unknown

Published: March 16, 2025

ABSTRACT Agricultural and environmental economists are in the fortunate position that a lot of what is happening on ground observable from space. Most agricultural production happens open one can see space when where innovations adopted, crop yields change, or forests converted to pastures, name just few examples. However, converting remotely sensed images into measurements particular variable not trivial, as there more pitfalls nuances than “meet eye”. Overall, however, research benefits tremendously advances available satellite data well complementary tools, such cloud‐based platforms, machine learning algorithms, econometric approaches. Our goal here provide with an accessible introduction working data, show‐case applications, discuss solutions, emphasize best practices. This supported by extensive supporting information, we describe how create different variables, common workflows, discussion required resources skills. Last but least, example reproducible codes made online.

Language: Английский

---

Monitoring road development in Congo Basin forests with multi-sensor satellite imagery and deep learning

Remote Sensing of Environment, Journal Year: 2024, Volume and Issue: unknown, P. 114380 - 114380

Published: Sept. 1, 2024

Language: Английский

---

A Weakly Supervised Semantic Segmentation Framework for Medium-resolution Forest Classification with Noisy Labels and GF-1 WFV Images

IEEE Transactions on Geoscience and Remote Sensing, Journal Year: 2024, Volume and Issue: 62, P. 1 - 19

Published: Jan. 1, 2024

Forests are the most widely distributed terrestrial vegetation type and play a significant role in global carbon cycle ecological diversity. Accurate timely forest detection provides essential data for management development. Current forest-related products differ definition, accuracy, spatial consistency, making them difficult to use. Therefore, it is necessary map cover under unified framework. However, detecting forests on large scale requires high-quality representative samples, which can be challenging. This study proposes weakly supervised classification framework (WSFCF) that uses noisy labels. The WSFCF designed address label generation, correction, sample location optimization. We employ spectral-spatial network extract accurately medium-resolution classification. experimental results show proposed method outperforms compared methods, achieving an accuracy of 91.76% OA 88.28% F1 score 110 GF-1 WFV images. supports subsequent extraction national-scale encourages mapping China's using Moreover, produces satisfactory outcomes objects such as water, farmland, built-up areas within area, demonstrating its effectiveness potential transferability.

Language: Английский

---

High-resolution sensors and deep learning models for tree resource monitoring

Nature Reviews Electrical Engineering, Journal Year: 2024, Volume and Issue: 2(1), P. 13 - 26

Published: Nov. 15, 2024

Language: Английский

---

To enhance sustainable development goal research, open up commercial satellite image archives

Proceedings of the National Academy of Sciences, Journal Year: 2025, Volume and Issue: 122(7)

Published: Feb. 12, 2025

The preference for simple explanations, known as the parsimony principle, has long guided development of scientific theories, hypotheses, and models. Yet recent years have seen a number successes in employing highly complex models ...

Language: Английский

---

Preface: Advancing deep learning for remote sensing time series data analysis

Remote Sensing of Environment, Journal Year: 2025, Volume and Issue: unknown, P. 114711 - 114711

Published: March 1, 2025

Language: Английский

---

Drivers and benefits of natural regeneration in tropical forests

Published: April 21, 2025

Language: Английский

---

Do finer-resolution sensors better discriminate burnt areas? A case study with MODIS, Landsat-8 and Sentinel-2 spectral indices for the Pantanal 2020 wildfire detection

International Journal of Remote Sensing, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 24

Published: April 29, 2025

Language: Английский

---