Spatio-temporal Prediction Research Articles

Improving Precipitation Forecasting through Early Fusion and Spatiotemporal Prediction: A Case Study Using the MultiPred Model

Due to the complexity and uncertainty of meteorological systems, traditional precipitation forecasting methods have certain limitations. Therefore, based on the common characteristics of meteorological data, a precipitation forecasting model named MultiPred is proposed, with the goal of continuously predicting precipitation for 4 h in a specific region. This model combines the multimodal fusion method with recursive spatiotemporal prediction models. The training and testing process of the model roughly involves using spatial feature extraction networks and temporal feature extraction networks to generate preliminary predictions for multimodal data. Subsequently, a modal fusion layer is employed to further extract and fuse the spatial features of the preliminary predictions from the previous step, outputting the predicted precipitation values for the target area. Experimental tests and training were conducted using ERA5 multi-meteorological modal data and GPM satellite precipitation data from 2017 to 2020, covering longitudes from 110° to 122° and latitudes from 20° to 32°. The training set used data from the first three years, while the validation set and test set each comprised 50% of the data from the fourth year. The initial learning rate for the experiment was set to 1 × 10−4, and training was performed for 1000 epochs. Additionally, the training process utilized a loss function composed of Mean Absolute Error (MAE), Mean Squared Error (MSE), and Structural Similarity Index (SSIM). The model was evaluated using the Critical Success Index (CSI), Probability of Detection (POD), and the Heidke Skill Score (HSS). Experimental results demonstrate that MultiPred excels in precipitation forecasting, particularly for light precipitation events with amounts greater than or equal to 0.1 mm and less than 2 mm. It achieves optimal performance in both light and heavy precipitation forecasting tasks.

Markovian Spatio-Temporal Pattern Variation and Prediction of Green Financial Development in China's Economic Belt Based on ADABOOST Algorithm

This paper constructs a comprehensive index system, measures Chinas provincial G-FINANCING INDEX using the VIKOR algorithm, explores the spatial and temporal distribution pattern of green financial development, hotspot regional migration and discrete trends by establishing a Markov spatial transfer matrix, and predicts the future pattern of green financial development in Chinas three major economic belts based on the machine-learning algorithm ADABOOST, which finds that: (1) the central region maintains a strong development momentum. The development of green finance in the region shows a scale effect, and there is a positive transmission trend of development momentum between cities. The right side of the nuclear density curve in the western region has a clear trailing trend, and the overall degree of improvement is high. The phenomenon of multi-polarisation exists among city clusters. The city clusters in the eastern region as a whole show extremely high development momentum. (2) Geographic background plays an important role in the process of transferring G-FINANCING INDEX in Chinese cities, and the level of regional G-FINANCING INDEX is the result of the joint action of desired outputs and non-desired outputs, which is dominated by the economic activities in the region, and is specifically expressed in the spatial spillover effect of regional G-FINANCING INDEX is the spatial spillover effect of regional economic activities. (3) The evolution of the G-FINANCING INDEX in the study region has obvious spatial spillover effects, and is synergistic with the type of regional G-FINANCING INDEX, showing the phenomenon of club convergence.

A spatiotemporal convolution recurrent neural network for pixel-level peripapillary atrophy prediction using sequential fundus images

The progression of peripapillary atrophy (PPA) is closely associated with the development of retinal diseases such as myopia and glaucoma. PPA prediction employing longitudinal images to obtain its progress trend can facilitate personalized treatment. Although existing studies have attempted to predict the persistence of PPA, such studies cannot provide quantitative measurement for personalized treatment. In this paper, we propose a spatiotemporal framework for pixel-level PPA prediction using sequential fundus images, including feature extractor, temporal memory, and spatiotemporal prediction modules. To take advantage of historical information, a temporal memory module is used, integrating current and prior features to build sequential data of features. To further enhance the prediction performance, the recurrent neural network states in a spatiotemporal prediction module transmit between different layers, enabling high-level states to guide the learning of low-level states. To handle missing data in clinical follow-up data, we use the predicted output of the spatiotemporal prediction module to substitute the missing data, and the scheduled-sampling strategy is employed in training. Extensive experiments conducted using a clinical dataset demonstrate that our proposed method achieves a satisfactory performance compared with the start-of-the-art models. The proposed approach can be applied using clinical data to obtain various quantitative indicators for personalized treatment and prevention of retinal disease.

Deep learning from latent spatiotemporal information of the heart: Identifying advanced bioimaging markers from echocardiograms.

A key strength of echocardiography lies in its integration of comprehensive spatiotemporal cardiac imaging data in real-time, to aid frontline or bedside patient risk stratification and management. Nonetheless, its acquisition, processing, and interpretation are known to all be subject to heterogeneity from its reliance on manual and subjective human tracings, which challenges workflow and protocol standardization and final interpretation accuracy. In the era of advanced computational power, utilization of machine learning algorithms for big data analytics in echocardiography promises reduction in cost, cognitive errors, and intra- and inter-observer variability. Novel spatiotemporal deep learning (DL) models allow the integration of temporal arm information based on unlabeled pixel echocardiographic data for convolution of an adaptive semantic spatiotemporal calibration to construct personalized 4D heart meshes, assess global and regional cardiac function, detect early valve pathology, and differentiate uncommon cardiovascular disorders. Meanwhile, data visualization on spatiotemporal DL prediction models helps extract latent temporal imaging features to develop advanced imaging biomarkers in early disease stages and advance our understanding of pathophysiology to support the development of personalized prevention or treatment strategies. Since portable echocardiograms have been increasingly used as point-of-care imaging tools to aid rural care delivery, the application of these new spatiotemporal DL techniques show the potentials in streamlining echocardiographic acquisition, processing, and data analysis to improve workflow standardization and efficiencies, and provide risk stratification and decision supporting tools in real-time, to prompt the building of new imaging diagnostic networks to enhance rural healthcare engagement.

A spatio-temporal temperature prediction model for coal spontaneous combustion based on back propagation neural network

Coal spontaneous combustion (CSC) occurs with the change of temperature and gas concentration which is useful for early warning. To explore the intrinsic correlation between temperature evolution and gas distribution, a Back Propagation Neural Network (BPNN) model of coal temperature - gas composition - spatial distance was constructed, which is based on the gas characteristics and spatial distribution of coal temperature in large-scale coal spontaneous combustion experiments. The results showed that the O2 concentration reached the lowest value (8.058%) at 45 cm on Day 52 with the time. CO and CO2 also formed concentration peaks at 45 cm, and the gas distribution continuously moved towards the inlet position. The hydrocarbon gas components showed periodic increases, with values below 100 ppm. When the supply air time increased, the temperature rise rate in the initial stage of oxidation was slow but relatively uniform. Affected by ventilation conditions and coal density, two high-temperature zones were observed at 45 cm and 85 cm in the experimental furnace. With the inclusion of spatial distance parameters in coal temperature prediction, the BPNN model exhibits accurate predictive capability, with a total error of 0.94997. This can provide a method for predicting temperature changes in coal mine.

A Bayesian approach using spatiotemporal features for suitable next hop selection in opportunistic networks

SummaryOpportunistic network (OppNet) belongs to the category of Mobile Ad‐hoc Networks (MANETs), a kind of Delay Tolerant Network (DTN), where the wireless nodes are completely mobile and the data transmission routes are dynamic. The major challenge in developing a routing model for such a network is the unpredictable nature of the movement of the nodes. In this paper, a spatiotemporal prediction model based on human mobility pattern is proposed using Bayesian posterior probability (BPPR) where several clusters are identified within the network and the day and time duration of nodes visiting those clusters are recorded. The Bayesian posterior probability is then used to determine the probability of the neighbor node visiting the destination's cluster. If the calculated probability for that node is higher than a specified threshold, the packet will be forwarded. A comparison of the results obtained on simulation is made with benchmark models—Epidemic, Prophet, HBPR, EDR, NexT, and EBC, to name a few, where it is found that on average the proposed model outperforms the other models in terms of delivery probability, hop count and number of messages dropped by around 23.89%, 24.8%, 24.4%, 37%, 11%, and 42% respectively, with varying number of nodes, TTL, message generation interval, and buffer size. Similar improvements have been observed in terms of the other two metrics. In terms of overhead ratio, the proposed model outperforms Epidemic, Prophet, HBPR, NexT, and EBC. However, as the number of nodes and TTL are varied, BPPR performs better than NexT by around 9% and 12%, respectively, in terms of average latency.

Research on Short-Term Prediction Methods for Small-Scale Three-Dimensional Wind Fields

The accurate prediction of small-scale three-dimensional wind fields is of great practical significance for aviation safety, wind power generation, and related fields. This study proposes a novel method for predicting small-scale three-dimensional wind fields by combining the mesoscale Weather Research and Forecasting (WRF) model with Computational Fluid Dynamics (CFD). The method consists of three components: the WRF module, the hybrid neural network prediction module, and the CFD module. First, mesoscale meteorological fields are simulated using the WRF module to establish a historical inflow boundary dataset for the CFD domain. Next, deep separable convolutions are incorporated, and convolutional long short-term memory (ConvLSTM) is combined with a deep separable convolution-gated recurrent unit (DSConvGRU) to construct a hybrid neural network prediction module named ConvLSTM-DSConvGRU. This module is employed for predicting inflow boundary data. Finally, the predicted inflow boundary conditions drive the CFD module to predict small-scale three-dimensional wind fields. The effectiveness of the WRF and CFD downscaling coupling method was validated using observed data from meteorological stations within the simulated domain, along with statistical indicators of errors. Additionally, a comparative evaluation was conducted between the proposed hybrid network model and the four commonly used spatiotemporal prediction models to assess its prediction performance. The results demonstrate that our proposed wind field prediction method achieves accurate simulation and short-term prediction of small-scale three-dimensional wind fields, and the hybrid network model exhibits comprehensive advantages in terms of model complexity and prediction accuracy.

MTAN: Multi-degree Tail-aware Attention Network for human motion prediction

Human motion prediction (HMP) aims to predict future human motions from historical pose sequences. Extensive efforts have adopted the Transformers or Graph Neural Networks (GNNs) to capture the spatio-temporal relationships between poses and thus incorporate the contextual information and complex behavior relationships for motion inference. However, most existing approaches treat the HMP task as a deterministic problem, thus resulting in poor diversity and long tail problems. This study attributes such issues to positional bias within the Transformers and the lack of degrees of freedom within the predictive model. Hence, we propose a novel Multi-degree Tail-aware Attention Network (MTAN) involving a tail-aware attention mechanism and a multi-degree feature representation strategy. Specifically, we introduce a tail-aware attention mechanism to adeptly capture spatio-temporal dependencies that accommodate both head and tail actions. Based on CVAE, the multi-degree feature representation strategy learns to capture temporal diversity by learning the joint distribution of observed and future sequences. Ultimately, we leverage GCN to model spatial dependencies effectively, culminating in a comprehensive spatiotemporal prediction model. We evaluate the effectiveness of our approach using three benchmark datasets, including Human3.6M, AMASS, and 3DPW. The results demonstrate that our approach surpasses state-of-the-art transformer methods, establishing its superiority in HMP.

Digital twin modeling and leak diagnosis of temperature and stress fields in LNG storage tanks

Temperature and stress serve as crucial indicators for monitoring the health of LNG storage tanks. To address the limitations of traditional point-based monitoring and transition to comprehensive digital monitoring, this paper introduces the application of digital twin (DT) technology. The DT model plays a pivotal role in accurately representing the state of the physical entity and serves as the foundation for monitoring services and meeting application requirements within DT systems. This article introduces a numerical simulation method guided by temperature sensor data. It utilizes a temperature and stress calculation program as a platform to create twin models for real-time assessment of the temperature and stress within the storage tank. By comparing with the Fluent simulation of physical field leakage, the DT model realizes real-time inversion of the temperature field and stress field from sensor data to the whole region. To enhance the diagnostic and predictive capabilities of the DT system, we flatten the computed temperature nephogram from the DT model and employ machine vision techniques to extract cloud map features. Machine learning methods such as support vector machine (SVM) and linear regression are utilized to achieve leakage diagnosis, leakage volume calculation, and leakage location calculation. Additionally, we introduce the sequence-to-sequence (Seq2seq) framework and combine it with convolutional neural networks (CNN) and long short-term memory (LSTM) to train the time series of nephogram, enabling spatiotemporal prediction of leakage areas.

A Deep U-Net-ConvLSTM Framework with Hydrodynamic Model for Basin-Scale Hydrodynamic Prediction

Traditional hydrodynamic models face the significant challenge of balancing the demands of long prediction spans and precise boundary conditions, large computational areas, and low computational costs when attempting to rapidly and accurately predict the nonlinear spatial and temporal characteristics of fluids at the basin scale. To tackle this obstacle, this study constructed a novel deep learning framework with a hydrodynamic model for the rapid spatiotemporal prediction of hydrodynamics at the basin scale, named U-Net-ConvLSTM. A validated high-fidelity hydrodynamic mechanistic model was utilized to build a 20-year hydrodynamic indicator dataset of the middle and lower reaches of the Han River for the training and validation of U-Net-ConvLSTM. The findings indicate that the R2 value of the model surpassed 0.99 when comparing the single-step prediction results with the target values. Additionally, the required computing time fell by 62.08% compared with the hydrodynamic model. The ablation tests demonstrate that the U-Net-ConvLSTM framework outperforms other frameworks in terms of accuracy for basin-scale hydrodynamic prediction. In the multi-step-ahead prediction scenarios, the prediction interval increased from 1 day to 5 days, while consistently maintaining an R2 value above 0.7, which demonstrates the effectiveness of the model in the missing boundary conditions scenario. In summary, the U-Net-ConvLSTM framework is capable of making precise spatiotemporal predictions in hydrodynamics, which may be considered a high-performance computational solution for predicting hydrodynamics at the basin scale.

An ultra-short-term wind power prediction method based on spatial-temporal attention graph convolutional model

Accurate and robust wind power forecasting (WPF) is great significance to ensure the safe and stable operation of the power system and promote the transformation of low-carbon energy. However, the high randomness and intermittency of wind power bring great challenges when designing reliable forecasting models. In this paper, a novel spatial-temporal attention graph convolutional network model is proposed. Firstly, the spatial attention mechanism is used to aggregate and extract the spatial correlations of the raw wind power data. Secondly, the temporal attention mechanism is applied to capture the temporal correlations. Then, the extracted spatial-temporal correlations were put into the temporal convolution network and the spatial convolution network to further obtain the temporal and spatial dependencies. Finally, the wind power forecasting results is output through the full connection layer. The proposed method is verified by using wind power data from real wind farm in China. The experimental results reveal that the proposed depth spatiotemporal prediction model has more significant advantages than other advanced models in terms of prediction accuracy and stability.

Leakage monitoring and diagnosis of LNG storage tanks with temperature sensing network integration and artificial intelligence algorithm

Temperature is the most important safety monitoring indicator for leakage diagnosis during the operation phase of liquefied natural gas (LNG) storage tanks. Timely monitoring and accurate identification of LNG leakage events are crucial for accident prevention, loss reduction, and facility safety maintenance. This study integrates artificial intelligence (AI) algorithms and temperature sensing data to achieve intelligent monitoring and diagnosis of leakage in LNG storage tanks. Firstly, a comprehensive temperature sensing network is constructed by combining numerical simulation of the temperature field and temperature sensing experiments using fiber Bragg grating sensors. Secondly, Python is used to perform linear grid interpolation and flattening on the sensing network, generating 2D temperature nephogram samples that are conducive to AI algorithm recognition. Finally, sample features are extracted using machine vision, and leakage location calculation, leakage diagnosis and leakage volume calculation are implemented with the help of machine learning algorithms, achieving satisfactory accuracy on the test set. In addition, the ConvLSTM framework is introduced for deep learning and recurrent neural network training, enabling spatiotemporal prediction of the leakage area.

STGWN: Enhanced spatiotemporal wave forecasting using multiscale features

This paper addresses the challenge of low accuracy in wave height prediction due to inadequate consideration of irregular topological structures and spatio-temporal dependencies of wave measurement points. We propose a novel spatio-temporal wave prediction method, named the Spatio-Temporal Graph Wave Network (STGWN). This model employs graph convolution instead of conventional 2D convolution, enabling spatial domain modeling of wave fields on a graph structure. Simultaneously, it utilizes LSTM for temporal modeling, effectively enhancing the representation of complex dynamic spatio-temporal correlations among wave nodes. Additionally, a dynamic wave position-aware mechanism is constructed to better identify feature variations among different wave nodes. Furthermore, a multi-scale feature fusion mechanism is designed within the model to prevent the risk of crucial temporal feature disappearance during the extraction of spatial dimension features. The proposed spatio-temporal method for Significant Wave Height (SWH) prediction is validated on a wave dataset comprising four different node distribution characteristics within the study area. Results demonstrate that the STGWN model outperforms comparative models in terms of prediction accuracy and stability across multiple evaluation metrics in the four experimental setups.

A Unified Spatio-Temporal Inference Network for Car-Sharing Serial Prediction.

Car-sharing systems require accurate demand prediction to ensure efficient resource allocation and scheduling decisions. However, developing precise predictive models for vehicle demand remains a challenging problem due to the complex spatio-temporal relationships. This paper introduces USTIN, the Unified Spatio-Temporal Inference Prediction Network, a novel neural network architecture for demand prediction. The model consists of three key components: a temporal feature unit, a spatial feature unit, and a spatio-temporal feature unit. The temporal unit utilizes historical demand data and comprises four layers, each corresponding to a different time scale (hourly, daily, weekly, and monthly). Meanwhile, the spatial unit incorporates contextual points of interest data to capture geographic demand factors around parking stations. Additionally, the spatio-temporal unit incorporates weather data to model the meteorological impacts across locations and time. We conducted extensive experiments on real-world car-sharing data. The proposed USTIN model demonstrated its ability to effectively learn intricate temporal, spatial, and spatiotemporal relationships, and outperformed existing state-of-the-art approaches. Moreover, we employed negative binomial regression with uncertainty to identify the most influential factors affecting car usage.

Corrigendum: Spatio-temporal variation and prediction of ecological quality based on remote sensing ecological index a case study of Zhanjiang City, China

Corrigendum: Spatio-temporal variation and prediction of ecological quality based on remote sensing ecological index a case study of Zhanjiang City, China

Short-term exposure to air pollution and hospital admission after COVID-19 in Catalonia: the COVAIR-CAT study.

A growing body of evidence has reported positive associations between long-term exposure to air pollution and poor COVID-19 outcomes. Inconsistent findings have been reported for short-term air pollution, mostly from ecological study designs. Using individual-level data, we studied the association between short-term variation in air pollutants [nitrogen dioxide (NO2), particulate matter with a diameter of <2.5 µm (PM2.5) and a diameter of <10 µm (PM10) and ozone (O3)] and hospital admission among individuals diagnosed with COVID-19. The COVAIR-CAT (Air pollution in relation to COVID-19 morbidity and mortality: a large population-based cohort study in Catalonia, Spain) cohort is a large population-based cohort in Catalonia, Spain including 240902 individuals diagnosed with COVID-19 in the primary care system from 1 March until 31 December 2020. Our outcome was hospitalization within 30 days of COVID-19 diagnosis. We used individual residential address to assign daily air-pollution exposure, estimated using machine-learning methods for spatiotemporal prediction. For each pandemic wave, we fitted Cox proportional-hazards models accounting for non-linear-distributed lagged exposure over the previous 7days. Results differed considerably by pandemic wave. During the second wave, an interquartile-range increase in cumulative weekly exposure to air pollution (lag0_7) was associated with a 12% increase (95% CI: 4% to 20%) in COVID-19 hospitalizations for NO2, 8% (95% CI: 1% to 16%) for PM2.5 and 9% (95% CI: 3% to 15%) for PM10. We observed consistent positive associations for same-day (lag0) exposure, whereas lag-specific associations beyond lag0 were generally not statistically significant. Our study suggests positive associations between NO2, PM2.5 and PM10 and hospitalization risk among individuals diagnosed with COVID-19 during the second wave. Cumulative hazard ratios were largely driven by exposure on the same day as hospitalization.

Dynamic predictive coding: A model of hierarchical sequence learning and prediction in the neocortex.

We introduce dynamic predictive coding, a hierarchical model of spatiotemporal prediction and sequence learning in the neocortex. The model assumes that higher cortical levels modulate the temporal dynamics of lower levels, correcting their predictions of dynamics using prediction errors. As a result, lower levels form representations that encode sequences at shorter timescales (e.g., a single step) while higher levels form representations that encode sequences at longer timescales (e.g., an entire sequence). We tested this model using a two-level neural network, where the top-down modulation creates low-dimensional combinations of a set of learned temporal dynamics to explain input sequences. When trained on natural videos, the lower-level model neurons developed space-time receptive fields similar to those of simple cells in the primary visual cortex while the higher-level responses spanned longer timescales, mimicking temporal response hierarchies in the cortex. Additionally, the network's hierarchical sequence representation exhibited both predictive and postdictive effects resembling those observed in visual motion processing in humans (e.g., in the flash-lag illusion). When coupled with an associative memory emulating the role of the hippocampus, the model allowed episodic memories to be stored and retrieved, supporting cue-triggered recall of an input sequence similar to activity recall in the visual cortex. When extended to three hierarchical levels, the model learned progressively more abstract temporal representations along the hierarchy. Taken together, our results suggest that cortical processing and learning of sequences can be interpreted as dynamic predictive coding based on a hierarchical spatiotemporal generative model of the visual world.

FBS-TGCN: a temporal graph-convolutional-network model for spatiotemporal prediction of spam messages from fake base stations

Spam messages affect the normal communication of mobile phone users and public security. Fake base station (FBS) is the main channel for sending spam messages. Accurate spatiotemporal prediction of spam messages from FBS is important for implementing short-term spam prevention measures by the police. However, existing models used for spam messages modeling often focus on capturing spatial dependencies based on distance, thus failing to fully exploit spatial dependencies hidden in the data. Meanwhile, they encounter difficulties in mitigating the impact of excessive zero data. Therefore, based on the crowdsourced spam message data, a temporal graph-convolutional-network model (FBS-TGCN) is developed for the spatiotemporal prediction of spam messages from FBS. This model includes graph convolutional network (GCN) and gated temporal convolutional network (TCN) to capture the spatial and temporal dependencies alternatively. We further combine a weighted adjacency matrix and a self-adaptive adjacency matrix to capture both distance-based spatial dependencies and hidden spatial dependencies. A weighted loss function is used to mitigate the influence of excessive zero data. Experimental results using the spam dataset in Beijing demonstrate the effectiveness of the proposed model when compared to the baseline models, especially in the prediction of spam messages quantity and high spamming areas.

Spatio-temporal sequence prediction of CO2 flooding and sequestration potential under geological and engineering uncertainties

CO2 injection for subsurface hydrocarbon development not only enhances oil and gas recovery but also enables CO2 sequestration in the subsurface. It is essential to develop effective methods for evaluating the potential of CO2 flooding and sequestration. Despite the existence of methods for predicting the effectiveness of hydrocarbon development using historical production data, insufficient emphasis has been placed on adequately incorporating geological and engineering uncertainty information to enhance prediction accuracy. To address this issue, a novel spatial-temporal ResNet (ST-ResNet) model is proposed for predicting hydrocarbon production, CO2 sequestration volume and CO2 diffusion pattern in the subsurface, which represent the CO2 flooding and sequestration potential. First, high-dimensional reservoir property fields are parameterized using the combined method of principal component analysis and discrete cosine transform (PCA-DCT). Second, the spatial sequence information of various reservoir property fields is extracted with features based on residual neural network (ResNet). Then, the time series information such as dynamic well control parameters is encoded with stacked BiLSTM (SBiLSTM). Specifically, the ST-ResNet model incorporates the above modules to overcome the limitations of collaborative consideration of temporal and spatial information involved in subsurface hydrocarbon development. Comparison between simulation and prediction results on the 2D/3D reservoir model reveals a significant achievement in prediction accuracy by the ST-ResNet model (with R2 and SSIM scores of 0.947, 0.911 and 0.937, 0.922, respectively). In comparison to CNN, LSTM and their combined approach CNN-LSTM, the ST-ResNet model demonstrates an improvement of 3% to 5% in R2, along with reductions of 20% to 30% in MAE and 15% to 25% in RMSE, respectively. These results highlight the superior stability and generalization of the ST-ResNet model. The contribution of this work is to provide a more accurate and efficient prediction tool to guide the integrated development of CO2 flooding and sequestration in subsurface hydrocarbon reservoirs, which facilitates decision-making processes for engineers.

Study on the Spatiotemporal Evolution and Influencing Factors of Forest Coverage Rate (FCR): A Case Study on Yunnan Province Based on Remote Sensing Image Interpretation

The study of the forest coverage rate (FCR) is related to the ecological environment and sustainable development goals (SDGs) of a region. In light of the lack of an organic integration method of “spatiotemporal evolution, correlation analysis, and change prediction” and the lack of a methodology that integrates methods of “remote sensing (RS) and GIS, multi-phase LUCC, and construction of econometric models” in the research methods at present, this study focus on Yunnan, a typical border province located in China with a relatively fragile “innate” ecological environment, as the research area. Based on the interpretation of land use/land cover (LULC) data retrieved from seven periods RS images (1990, 1995, 2000, 2005, 2010, 2015, and 2020), the spatiotemporal evolution of FCR in 129 counties was analyzed. Complementary research methods, such as the spatial econometric model, geographically weighted regression (GWR), and the geographic detector (GD), are used to reveal the influencing factors of FCR. Finally, this study predicts the FCRs of 129 counties in Yunnan from 2025 to 2050. The FCR in Yunnan presents an increasing trend year by year, increasing from 28.96% in 1990 to 49.05% in 2020. In addition, it exhibits spatial agglomeration characteristics with fewer values in the east and more in the west. The analysis of influencing factors show that the increases in the per capita GDP, land utilization rate, and annual average temperature, and the implementation of the Conversion of Cultivated Land into Forest Project (CCFP) will significantly improve the FCR, while the increases in the population density land reclamation rate, the proportion of construction land area, and the proportion of soil erosion land area will significantly reduce the FCR. Furthermore, the FCR is influenced by multiple factors, and the relative factors observed not only show significant spatial differences, but also present complex and diverse patterns, with the additional characteristics of being interwoven and overlapping. This study contributes to expanding and improving the methods and pathways of exploring the spatiotemporal evolution characteristics of FCR in ecologically fragile areas using RS methods, providing a reference for increasing FCR and improving the ecological environment’s quality in Yunnan Province and other ecologically fragile areas.