Renyi Liu

Accurate, scalable estimation of rice planting dates is essential for climate-adaptive management in multi-cropping regions, yet most models rely on static calendars, which fail to capture climate-driven shifts and bias simulated yield responses. This study aims to develop a climate-driven, spatially explicit framework to simulate dynamic transplanting dates across diverse multi-cropping systems in monsoon Asia. Utilizing daily AgERA5 reanalysis and Monsoon Asia Rice Calendar (MARC) data from 2019 to 2020, we present Geo-ROCKET. The framework integrates an automated K-means clustering workflow to delineate bimodal planting windows and employs random convolutional kernel transforms with adaptive geographic neighborhoods to capture local climate heterogeneity. Evaluated by area-weighted mean absolute error (MAE), the model achieves high accuracy across six seasons (MAE 6.53–12.50 days), outperforming six traditional ROCKET and ensemble baselines while preserving smooth spatial error fields. Sensitivity experiments reveal that a 15-day bias in the previous harvest date can increase transplanting error to 10.8–17.8 days, emphasizing the importance of sequential consistency. By providing dynamic, climate-sensitive inputs, Geo-ROCKET improves the accuracy of crop modeling for climate impact projections. This framework offers a flexible tool for characterizing human management decisions and evaluating adaptation strategies in intensive agricultural systems.

Since the outbreak of the Russia–Ukraine conflict in 2022, Ukraine’s agricultural production has faced significant disruption, leading to widespread cropland abandonment. These croplands were abandoned at different stages, primarily due to war-related destruction and displacement of people. Existing methods for detecting abandoned cropland fail to account for crop type differences and distinguish abandonment stages, leading to inaccuracies. Therefore, this study proposes a novel framework combining crop-type classification with the Bias-weighted Time-Weighted Dynamic Time Warping (BTWDTW) method, distinguishing between sowing and harvest abandonment. Additionally, the proposed framework improves accuracy by integrating a more nuanced analysis of crop-specific patterns, thus offering more precise insights into abandonment dynamics. The overall accuracy of the proposed method reached 88.9%. The results reveal a V-shaped trajectory of cropland abandonment, with abandoned areas increasing from 28,184 km² in 2022 to 33,278 km² in 2024, with 2023 showing an abandoned area of 24,007.65 km². Spatially, about 70% of sowing abandonment occurred in high-conflict areas, with hotspots of unplanted abandonment shifting from southern Ukraine to the northeast, while unharvested abandonment was observed across the entire country. Significant variations were found across crop types, with maize experiencing the highest rate of unharvested abandonment, while wheat exhibited a more balanced pattern of sowing and harvest losses. The proposed method and results provide valuable insights for post-conflict agricultural recovery and decision-making in recovery planning.

Land surface temperature (LST) has a wide application in Earth Science-related fields, and spatial downscaling is an important method to retrieve high-resolution LST data. However, existing LST downscaling methods have difficulties in simultaneously constructing and expressing spatial non-stationarity, spatial autocorrelation, and complex non-linearity during the LST downscaling process, which limits the performance of the models. Moreover, there is a lack of research on high-resolution nighttime land surface temperature (NLST) reconstruction based on spatial downscaling, which does not meet the data needs for urban-scale nighttime urban heat island (UHI) studies. Therefore, this study combined Geographically Neural Network Weighted Regression (GNNWR) with Area-to-Point Kriging interpolation (ATPK) to propose a Geographically Neural Network Weighted Regression Kriging (GNNWRK) model for NLST downscaling. To verify the model’s generality and robustness, this study selected four study areas with different landform and climate type for NLST spatial downscaling experiments. The GNNWRK was compared with four benchmark downscaling methods, including TsHARP, Random Forest, Geographically Weighted Regression, and GNNWR. The results show that compared to these four benchmark methods, the GNNWRK method has higher accuracy in NLST downscaling, with a maximum Pearson’s Correlation Coefficient (Pcc) of 0.930 and a minimum Root Mean Square Error (RMSE) of 0.886 K. Moreover, the validation based on MODIS NLST data and ground-measured NLST data also indicates that the GNNWRK model can obtain more accurate, high-resolution NLST with richer and more detailed texture. This enhances the potential of NLST in studying the effects of urban nighttime heat islands at a finer scale.

Historical news media reports serve as a vital data source for understanding the risk of urban ground collapse (UGC) events. At present, the application of large language models (LLMs) offers unprecedented opportunities to effectively extract UGC events and their spatiotemporal information from a vast amount of news reports and media data. Therefore, this study proposes an LLM-based inventory construction framework consisting of three steps: news reports crawling, UGC event recognition, and event attribute extraction. Focusing on Zhejiang province, China, as the test region, a total of 27 cases of collapse events from 637 news reports were collected for 11 prefecture-level cities. The method achieved a recall rate of over 60% and a precision below 35%, indicating its potential for effectively and automatically screening collapse events; however, the accuracy needs to be improved to account for confusion with other urban collapse events, such as bridge collapses. The obtained UGC event inventory is the first open access inventory based on internet news reports, event dates and locations, and collapse co-ordinates derived from unstructured contents. Furthermore, this study provides insights into the spatial pattern of UGC frequency in Zhejiang province, effectively supplementing the statistical data provided by the local government.

Sea surface temperature (SST) is a key factor in the marine environment, and its accurate forecasting is important for climatic research, ecological preservation, and economic progression. Existing methods mostly rely on convolutional networks, which encounter difficulties in encoding irregular data. In this paper, allowing for comprehensive encoding of irregular data containing land and islands, we construct a graph structure to represent SST data and propose a graph memory neural network (GMNN). The GMNN includes a graph encoder built upon the iterative graph neural network (GNN) idea to extract spatial relationships within SST data. It not only considers node but also edge information, thereby adequately characterizing spatial correlations. Then, a long short-term memory (LSTM) network is used to capture temporal dynamics in the SST variation process. We choose the data from the Northwest Pacific Ocean to validate GMNN’s effectiveness for SST prediction in different partitions, time scales, and prediction steps. The results show that our model has better performance for both complete and incomplete sea areas compared to other models.

The spatial learned index constructs a spatial index by learning the spatial distribution, which performs a lower cost of storage and query than the spatial indices. The current update strategies of spatial learned indices can only solve limited updates at the cost of query performance. We propose a novel spatial learned index structure based on a Block Range Index (SLBRIN for short). Its core idea is to cooperate history range and current range to satisfy a fast spatial query and efficient index update simultaneously. SLBRIN deconstructs the update transaction into three parallel operations and optimizes them based on the temporal proximity of spatial distribution. SLBRIN also provides the spatial query strategy with the spatial learned index and spatial location code, including point query, range query and kNN query. Experiments on synthetic and real datasets demonstrate that SLBRIN clearly outperforms traditional spatial indices and state-of-the-art spatial learned indices in the cost of storage and query. Moreover, in the simulated real-time update scenario, SLBRIN has the faster and more stable query performance while satisfying efficient updates.

The collapse of buildings caused by the earthquake seriously threatened human lives and safety. So, the quick detection of collapsed buildings from post-earthquake images is essential for disaster relief and disaster damage assessment. Compared with the traditional building extraction methods, the methods based on convolutional neural networks perform better because it can automatically extract high-dimensional abstract features from images. However, there are still many problems with deep learning in the extraction of collapsed buildings. For example, due to the complex scenes after the earthquake, the collapsed buildings are easily confused with the background, so it is difficult to fully use the multiple features extracted by collapsed buildings, which leads to time consumption and low accuracy of collapsed buildings extraction when training the model. In addition, model training is prone to overfitting, which reduces the performance of model migration. This paper proposes to use the improved classic version of the you only look once model (YOLOv4) to detect collapsed buildings from the post-earthquake aerial images. Specifically, the k-means algorithm is used to optimally select the number and size of anchors from the image. We replace the Resblock in CSPDarkNet53 in YOLOv4 with the ResNext block to improve the backbone’s ability and the performance of classification. Furthermore, to replace the loss function of YOLOv4 with the Focal-EOIU loss function. The result shows that compared with the original YOLOv4 model, our proposed method can extract collapsed buildings more accurately. The AP (average precision) increased from 88.23% to 93.76%. The detection speed reached 32.7 f/s. Our method not only improves the accuracy but also enhances the detection speed of the collapsed buildings. Moreover, providing a basis for the detection of large-scale collapsed buildings in the future.

Hand, foot, and mouth disease (HFMD) is an epidemic infectious disease in China. Its incidence is affected by a variety of natural environmental and socioeconomic factors, and its transmission has strong seasonal and spatial heterogeneity. To quantify the spatial relationship between the incidence of HFMD (I-HFMD) and eight potential risk factors (temperature, humidity, precipitation, wind speed, air pressure, altitude, child population density, and per capita GDP) on the Chinese mainland, we established a geographically weighted regression (GWR) model to analyze their impacts in different seasons and provinces. The GWR model successfully describes the spatial changes of the influence of potential risks, and shows greatly improved estimation performance compared with the ordinary linear regression (OLR) method. Our findings help to understand the seasonally and spatially relevant effects of natural environmental and socioeconomic factors on the I-HFMD, and can provide information to be used to develop effective prevention strategies against HFMD at different locations and in different seasons.

The increase in atmospheric pollution dominated by particles with an aerodynamic diameter smaller than 2.5 μm (PM_2.5) has become one of the most serious environmental hazards worldwide. The geographically weighted regression (GWR) model is a vital method to estimate the spatial distribution of the ground-level PM_2.5 concentration. Wind information reflects the directional dependence of the spatial distribution, which can be abstracted as a combination of spatial and directional non-stationarity components. In this paper, a GWR model considering directional non-stationarity (GDWR) is proposed. To assess the efficacy of our method, monthly PM_2.5 concentration estimation was carried out as a case study from March 2015 to February 2016 in the Yangtze River Delta region. The results indicate that the GDWR model attained the best fitting effect (0.79) and the smallest error fluctuation, the ordinary least squares (OLS) (0.589) fitting effect was the worst, and the GWR (0.72) and directionally weighted regression (DWR) (0.74) fitting effects were moderate. A non-stationarity hypothesis test was performed to confirm directional non-stationarity. The distribution of the PM_2.5 concentration in the Yangtze River Delta is also discussed here.

Infrared observation is an all-weather, real-time, large-scale precipitation observation method with high spatio-temporal resolution. A high-precision deep learning algorithm of infrared precipitation estimation can provide powerful data support for precipitation nowcasting and other hydrological studies with high timeliness requirements. The “classification-estimation” two-stage framework is widely used for balancing the data distribution in precipitation estimation algorithms, but still has the error accumulation issue due to its simple series-wound combination mode. In this paper, we propose a multi-task collaboration framework (MTCF), i.e., a novel combination mode of the classification and estimation model, which alleviates the error accumulation and retains the ability to improve the data balance. Specifically, we design a novel positive information feedback loop composed of a consistency constraint mechanism, which largely improves the information abundance and the prediction accuracy of the classification branch, and a cross-branch interaction module (CBIM), which realizes the soft feature transformation between branches via the soft spatial attention mechanism. In addition, we also model and analyze the importance of the input infrared bands, which lay a foundation for further optimizing the input and improving the generalization of the model on other infrared data. Extensive experiments based on Himawari-8 demonstrate that compared with the baseline model, our MTCF obtains a significant improvement by 3.2%, 3.71%, 5.13%, 4.04% in F1-score when the precipitation intensity is 0.5, 2, 5, 10 mm/h, respectively. Moreover, it also has a satisfactory performance in identifying precipitation spatial distribution details and small-scale precipitation, and strong stability to the extreme-precipitation of typhoons.