The Pearl River Delta (PRD) is one of the most dynamic economic regions in the Asia-Pacific region. At present, it still faces many problems, such as the over-exploitation of urban land and unbalanced development. Through the study of the spatial–temporal characteristics of the development intensity of the PRD region and its driving factors, the key points and difficulties of urban development can be intuitively found. In previous studies, geodetector was widely used to determine the impact of driving factors. This paper uses several different research methods, including the Moran index, the semi-variability index, hot and cold spots, etc., based on the land use data of the PRD region in 1990, 2000, 2010, and 2020 to analyze the spatial–temporal characteristics of the development intensity. Combined with the socio-economic data of the statistical yearbook, factor detection and interaction detection of the 10 driving factors of development intensity are carried out based on geodetector, and reasonable optimization suggestions are put forward for the current situation of the region. The main conclusions are as follows: (1) The overall development intensity of the PRD region shows an upward trend, showing a “core periphery” spatial pattern of high center and low periphery centered around the Pearl River estuary. (2) The spatial distribution of cold and hot spots shows agglomeration, mainly in the form of high aggregation and low aggregation. (3) The driving factors for the development intensity for the PRD region in the past 30 years mainly include population agglomeration level, industrial structure level, economic strength level, terrain slope, etc. Among them, any two factors have a stronger interaction than a single factor, and all are enhanced by two factors. The dominant factors of interaction in different periods are different.

Accurate classification of forest tree species holds great significance in the context of forest biodiversity assessment and the management of forest resources. In this study, we utilized Sentinel-2 time series data with high temporal and spatial resolution for tree species classification. To address potential classification errors stemming from spectral differences due to tree age variations, we implemented the Continuous Change Detection and Classification (CCDC) algorithm to estimate tree ages, which were integrated as additional features into our classification models. Four different combinations of classification features were created for both the random forest (RF) algorithm and extreme gradient boosting (XGB) algorithm: spectral band (Spec), spectral band combined with tree age feature (SpecAge), spectral band combined with spectral index (SpecVI), and spectral band combined with spectral index and tree age feature (SpecVIAge). The results demonstrated that the XGB-based models outperformed the RF-based ones, with the SpecVIAge model achieving the highest accuracy at 78.8%. The incorporation of tree age as a classification feature led to an improvement in accuracy by 2% to 3%. The improvement effect on classification accuracy varies across tree species, due to the varying uniformity of tree age among different tree species. These results also showed it is feasible to accurately map regional tree species based on a time-series multi-feature tree species classification model which takes into account tree age.

The persistent increase in forest pest outbreaks requires timely detection methods to monitor the disaster precisely. However, early detection is challenging due to insufficient temporal observation and subtle tree changes. This article proposed a novel framework that collaborates multi-source remote sensing data and uses a change detection algorithm to archive early detection of infestation caused by Dendrolimus tabulaeformis Tsai et Liu (D. tabulaeformis) attacks. First, all available Sentinel-2 images with less than 20% cloud cover were utilized. During periods with long intervals (>16 days) between Sentinel-2 images, Landsat-8 images with less than 20% cloud cover were downscaled to a spatial resolution of 10 m using a deep learning algorithm to meet the requirement for a high temporal frequency of clear observations. Second, the spectral index differences between healthy and infested trees were examined to address the challenge of detecting subtle changes in pest attacks. The Enhanced Vegetation Index (EVI) was selected for early defoliation detection. On this basis, the EWMACD (Exponentially Weighted Moving Average Change Detection) algorithm, which is sensitive to subtle changes, was enhanced to improve the capability of detecting early D. tabulaeformis attacks. The assessment showed that the overall accuracy of the change detection (F1 score) reached 0.86 during the early stage and 0.88 during the late stage. The temporal accuracy (Precision) was 84.1% during the early stage. The accuracy significantly improved compared to using a single remote sensing data source. This study presents a new framework capable of monitoring early forest defoliation caused by D. tabulaeformis attacks and offering opportunities for predicting future outbreaks and implementing preventive measures.

Accurate information on the location of dominant tree species is essential for scientific forest management. However, factors like changes in forest phenology, stand conditions, and mixed understory backgrounds introduce uncertainties in remote sensing-based species mapping. To address these challenges, this study maps dominant tree species using time series Sentinel-2 data combined with environmental context data. To quantify the impact of understory background on mapping accuracy, this study applied a random forest inversion model to estimate the canopy cover across the study area. Binary contour plots and Pearson’s correlation coefficient were used to quantify the relationship between canopy cover and classification uncertainty at both the grid and pixels. A 10 m resolution map of dominant tree species in Yunnan Province, featuring eight species, was produced with an overall accuracy of 83.52% and a Kappa coefficient of 0.8115. The R² value between the predicted and actual tree area proportions was greater than 0.93, with RMSEs consistently below 2.6. In addition, we observed strong negative correlations between different canopy cover classes. The correlations were −0.67 for low-cover areas, −0.40 for medium-cover areas, and −0.73 for high-cover areas. Our mapping framework enables the accurate identification of regional dominant species, and the established relationship between understory context and classification uncertainty provides valuable insights for analyzing potential mapping errors.

Remote sensing change captioning aims to describe disaster-related changes from bi-temporal imagery, yet existing methods typically produce image-level captions without explicit regional evidence, limiting interpretability and weakening the link between generated language and actual changed areas. We present ChangePixel, a single-backbone framework that upgrades remote sensing change captioning into pixel-level, evidence-grounded change narration without introducing new manual grounding labels. ChangePixel incorporates three lightweight modules: a Bi-Temporal Change-Aware Transfer Adapter (BCTA) that converts shared pre- and post-event visual features into change-aware grounding representations, a Change Region Grounding Planner (CRGP) that localizes a compact set of informative changed regions before narration begins, and a Weak Evidence Alignment Bridge (WAB) that converts released change captions into phrase-to-region weak supervision. Through this design, the model jointly produces a global change caption and region-level evidence in the form of pixel masks paired with corresponding local change phrases. Experiments on the Remote Sensing Change Caption (RSCC) dataset and LEVIR-CC demonstrate that ChangePixel provides caption quality (ROUGE 19.52/ST5-SCS 76.91 on RSCC; CIDEr-D 56.82 on LEVIR-CC under zero-shot transfer) that is competitive with general-purpose vision–language models (VLMs) while adding pixel-level spatial evidence to change narration; additionally, evidence localization is quantified on LEVIR-MCI through semantic change-mask metrics, reaching Change mIoU 33.8 (15.3 points higher than a non-learned pixel-difference floor of 18.5), whereas phrase-to-region alignment is assessed qualitatively pending a dedicated grounding benchmark. The proposed framework offers a practical path from coarse image-level captioning to evidence-grounded disaster understanding.