Eduardo Eiji Maeda

Forest fragmentation has produced widespread edge zones, leading to declines in ecosystem functioning. Although human activities are the primary drivers of such ecological disturbance, how they modulate the depth and magnitude of edge effects remains insufficiently understood. Here we assessed edge effects in fragmented forests across China, a megadiversity nation with the largest afforested area and high deforestation risk. By 2023, nearly 88% of China’s forests are within 1 km of an edge, with roughly 95% of these edges adjoining anthropogenic land uses. Compared to natural forest edges, anthropogenic edges exhibit stronger edge effects on ecosystem productivity and its temporal stability, with productivity disrupted and stability declining markedly within ~210 m of edges. In contrast, natural edges show reduced productivity within ~100 m of edges but no detectable impact on ecosystem stability. These differences are primarily driven by canopy height, as well as changes in hydrological, thermal, and wind conditions. These findings underscore the importance of forest-edge conservation in China.

Climate change is altering vegetation patterns on the Qinghai-Tibet Plateau, where diverse Köppen climate zones reveal complex ecological responses. Here, we advance from a categorical view of climate, represented by Köppen-Geiger classifications, to a continuous characterization of the climate heterogeneity index to better capture the link between climate and vegetation across 1- and 10-km scales. We show that climate heterogeneity strengthens vegetation-climate relationships across spatial scales, with its influence increasing 1.2-fold at coarse resolution compared to fine scales. This enhancement stems from this index providing critical heterogeneity information absent in coarse macroclimate data, thereby refining suitability predictions in regions where fine-scale topographic variability is unresolved. At finer 1-km scales, topographic heterogeneity inherently dominates, reducing the relative impact of climate heterogeneity. Overall, incorporating climate heterogeneity compensates for coarse-scale data limitations, enhancing suitability predictions where topographic detail is lost, thereby bridging macroclimate biases and microscale dynamics, and advancing robust ecological forecasting.

Rainfall among the Amazon has been shown to decline sharply once forest loss exceeds a critical threshold, typically around 30%–40%. However, future interactions between climate change and deforestation on this threshold remain unclear. We model their combined effects by 2050 on rainfall and temperature. Our findings indicate that climate and land‐use change lower the deforestation threshold for rainfall decline—from 50% forest loss under land‐use change alone to 45% under SSP1‐2.6 and 10% under SSP5‐8.5 (90 × 90 km reference grids). Land‐use change alone is projected to warm the region by 0.29°C, while the combined effects under SSP5‐8.5 lead to 1.4°C warming. Annual rainfall is projected to decline by 1.7% due to land‐use change alone and up to 13.9% under SSP1‐2.6. Future climates amplify the importance of forests in maintaining moisture supply for rainfall. These findings underscore the growing vulnerability of the Southern Amazon to rainfall disruption and reduced land–atmosphere system resilience.

Forest microclimates regulate biodiversity, ecosystem functioning, and species' thermal tolerance, yet tropical forest disturbances (logging, fire and fragmentation) increasingly disrupt these conditions. Southeast Asia, home to 15% of the world's tropical forests, faces rapid deforestation and degradation, but large‐scale changes in maximum microclimate temperature (Tmicro _max ) remain poorly quantified. Here, we estimated baseline average Tmicro _max (1984–2014), measured 15 cm above the surface and projected changes by 2050 under different scenarios combining climate forcing and forest cover change. Baseline Tmicro _max ranged from 11.6°C to 25.1°C, driven mainly by topography, macroclimate, and vegetation structure. Projections indicate regional mean increases of 1.4°C (SSP2‐4.5) to 2.1°C (SSP5‐8.5). By 2050%, 52%–66% of forests may experience Tmicro _max values warmer than any in the baseline period on average with Brunei, Indonesia, and Malaysia showing the strongest changes. These results reveal substantial sub‐canopy warming, posing potential risks for tropical biodiversity and highlighting the need for spatially informed conservation planning.

Montane forests are biodiversity hotspots that provide important ecosystem services, including temperature buffering for numerous species underneath forest canopies. In recent decades, montane forests have been under increasing pressure from small-and large-scale deforestation, yet associated spatio-temporal changes in temperature buffering capacity remain unclear. Here, we studied the changes in temperature buffering capacity due to forest loss from 2003 to 2022 in three montane forest ecosystems in Africa (Mount Kilimanjaro, Mount Bale, and the Taita Hills). We modeled the temperature buffering changes based on in situ microclimate measurements inside forests and in open areas, climate data, airborne laser scanning data, and satellite observations. We found that during the study period montane forests were lost at a rate of 2–9% across the study areas. This loss led to an annual average microclimate air temperature warming ranging from 2.0 ± 0.8 °C to 5.6 ± 2.1 °C across the three montane forests. The warming reduced the maximum air temperature buffering by an average of 3 ± 1.5 °C. Locally, the temperature buffering disappeared over time and transitioned to a mesoclimate amplification. Our findings demonstrate that microclimate buffering capacity was markedly diminished as a result of microclimate warming driven by recent forest loss.

Timely and accurate wildfire detection is critical for ecological monitoring and disaster response. Using a model to handle both long-term time series and multi-scale spatial features is challenging. In this study, we propose a novel burned area detection algorithm based on the CL-UNet (ConvLstm-U-Net) model. CL-UNet integrates convolutional long short-term memory (ConvLSTM) units with a multi-scale encoder-decoder architecture to capture both spatial and temporal dependencies inherent in satellite image time series. The model is trained to forecast spectral index values for the target year, and discrepancies between the predicted and observed values are used to identify anomalous changes. Burned areas are detected using a root mean square error (RMSE)-based thresholding method, which provides a robust mathematical foundation for selecting the optimal threshold. Using MODIS products with a temporal coverage spanning up to 9 years, we constructed spectral index time series (SITS) based on multiple spectral indices. Experimental results from two large-scale forest fire demonstrate that Burn Area Index (BAI) exhibits the most qualified and stable performance. Under various conditions, the Recall, Precision and F1 score of the model using the BAI time series as input data all exceed 80%. The results further indicate that CL-UNet achieves outstanding performance in both study areas. When utilizing the BAI time series as input, the Recall, Precision and F1 score of CL-UNet are 1% to 10% higher than those of other models. The proposed algorithm shows strong potential for general spatiotemporal disturbance detection tasks in remote sensing applications beyond wildfire monitoring.

Although diverse climates shape complex vegetation distribution globally, how climate heterogeneity regulates the impact of climate change on vegetation migration remains unclear. In this study, we used high‐resolution Köppen‐Geiger climate classifications to assess climate heterogeneity across the Qinghai‐Tibet Plateau from 1990 to 2020. We found that areas with higher climate heterogeneity support greater vegetation adaptability and migration potential in response to changing climate conditions. Including climate heterogeneity in species distribution models clearly confirmed that climate heterogeneity promotes vegetation migration. Our results show that while rising temperatures generally limit vegetation migration, climate heterogeneity can help offset this effect by offering climatic refugia and expanding suitable habitat ranges. The interaction between climate heterogeneity and temperature increase had a positive moderating effect. By incorporating this factor, our approach provides a more accurate understanding of vegetation dynamics and supports the development of adaptive conservation strategies for biodiversity.