Exploring spatiotemporal dynamics of NDVI and climate-driven responses in ecosystems: Insights for sustainable management and climate resilience

Ecological Informatics, Год журнала: 2024, Номер 80, С. 102532 - 102532

Опубликована: Фев. 22, 2024

Understanding the intricate relationship between climate variables and Normalized Difference Vegetation Index (NDVI) is essential for effective ecosystem management. This study focuses on spatiotemporal dynamics of NDVI its interaction with in ecologically diverse Khyber Pakhtunkhwa (KPK) Province, Pakistan, from 2000 to 2022. The research methodology involves analyzing satellite images meteorological datasets examine surface latent heat flux (SHF), total precipitation (TPP), temperature (T), soil (ST), pressure (TP). KPK Province's ecological significance complex climate-vegetation interactions drive selection this area. uses multiple linear regression analysis investigate how T, TPP, SHF, TP influence NDVI. Mann-Kendall test detects trends, Sen's slope estimator quantifying trend magnitudes. Additionally, correlation coefficients provide insights into long-term changes association strengths. findings highlight a consistent upward mean over 23 years, revealing an overall increase NDVI, particularly vegetation-dense areas where it rose 0.27 0.32. showed annual growth rate 0.84% entire area, specific vegetated zones exhibiting slightly lower 0.80%. However, average yearly higher vegetation-specific (0.00237) compared whole area (0.00151). occurs alongside statistically significant decrease SHF PPT, suggesting adaptation vegetation changing conditions Province. In contrast, exhibits negative −5.952e-06 (p < 0.05), indicating pronounced downward trend. Similarly, estimate demonstrates −0.0001 showing diminishing precipitation. uncovers linkages within These have far-reaching implications, guiding decision-making land management, conservation efforts, global resilience strategies. Ultimately, underscores critical role data-driven approaches shaping greener more sustainable future.

Язык: Английский

---

Remote sensing of terrestrial gross primary productivity: a review of advances in theoretical foundation, key parameters and methods

GIScience & Remote Sensing, Год журнала: 2024, Номер 61(1)

Опубликована: Фев. 20, 2024

Accurately estimating gross primary productivity (GPP), the largest carbon flux in terrestrial ecosystems, is crucial for advancing our understanding of global cycle and predicting climate feedbacks. The advancements remote sensing (RS) have facilitated development GPP estimation models at regional scales recent decades. This article systemically reviews RS-based three main aspects: theoretical foundation, key parameters methods. Regarding RS generally excels representing characteristics during light transmission process photosynthesis. However, it exhibits a relatively weaker ability to describe reaction process, severely limiting in-depth mechanisms estimation. Concerning parameters, definition traditional such as leaf area index (LAI), photosynthetically active radiation (PAR), fraction absorbing PAR, has been detailed (e.g. LAI divided into sunlit shaded LAI). their accuracy still needs improvement. Additionally, researchers developed effective photochemical reflectance index, sun-induced chlorophyll fluorescence, maximum carboxylation rate) that possess increased capability represent interpret methods, although four categories (statistical model, use efficiency model machine learning-based model) made significant progress parameter optimization, mechanism innovation remain less than satisfactory. Finally, we summarize current issues related performance accuracy, capabilities, well scale connotation mismatch. Integrating more adequate situ comprehensive observations would enhance interpretability models, providing reliable insights future studies. contributes photosynthetic estimation, potentially aiding optimization (improving existing developing new ones) design (introducing exploring mechanistic models).

Язык: Английский

---

Artificial intelligence for modeling and understanding extreme weather and climate events

Nature Communications, Год журнала: 2025, Номер 16(1)

Опубликована: Фев. 24, 2025

In recent years, artificial intelligence (AI) has deeply impacted various fields, including Earth system sciences, by improving weather forecasting, model emulation, parameter estimation, and the prediction of extreme events. The latter comes with specific challenges, such as developing accurate predictors from noisy, heterogeneous, small sample sizes data limited annotations. This paper reviews how AI is being used to analyze climate events (like floods, droughts, wildfires, heatwaves), highlighting importance creating accurate, transparent, reliable models. We discuss hurdles dealing data, integrating real-time information, deploying understandable models, all crucial steps for gaining stakeholder trust meeting regulatory needs. provide an overview can help identify explain more effectively, disaster response communication. emphasize need collaboration across different fields create solutions that are practical, understandable, trustworthy enhance readiness risk reduction. Artificial Intelligence transforming study like helping overcome challenges integration. review article highlights models improve response, communication trust.

Язык: Английский

---

Developing machine learning models for wheat yield prediction using ground-based data, satellite-based actual evapotranspiration and vegetation indices

European Journal of Agronomy, Год журнала: 2023, Номер 146, С. 126820 - 126820

Опубликована: Апрель 2, 2023

Язык: Английский

---

Monthly NDVI Prediction Using Spatial Autocorrelation and Nonlocal Attention Networks

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, Год журнала: 2024, Номер 17, С. 3425 - 3437

Опубликована: Янв. 1, 2024

Accurate prediction of vegetation indices is useful for helping maintain stability, sustaining food production, and reducing socioeconomic losses. The traditional convolutional long short-term memory (ConvLSTM) model ignores the spatial aggregation characteristics normalized difference index (NDVI) itself global dependence information in space. In this study, we propose a new NDVI method, namely, ConvLSTM with autocorrelation nonlocal attention module (ConvLSTM-SAC-NL), by combining to capture long-range modeling based on local Moran learn dependence. experimental results indicate that ConvLSTM-SAC-NL outperforms seven baseline forecasting models, an R ² 0.881 monthly Huangpi District Wuhan City, relative values 0.758, 0.777, 0.741, 0.776, 0.804, 0.829 0.815 random forest (RF), support vector machine regression (SVR), shortterm (LSTM), bidirectional (BiLSTM), graph network (GCN), predictive recurrent neural (PredRNN) respectively. Spatially, demonstrate improved accuracy over 91.49% study area when compared ConvLTSM. Therefore, proposed could serve as effective approach conditions at regional scales.

Язык: Английский

---

Enhancing the accuracy of monitoring effective tiller counts of wheat using multi-source data and machine learning derived from consumer drones

Computers and Electronics in Agriculture, Год журнала: 2025, Номер 232, С. 110120 - 110120

Опубликована: Фев. 24, 2025

Язык: Английский

---

Veg-W2TCN: A parallel hybrid forecasting framework for non-stationary time series using wavelet and temporal convolution network model

Applied Soft Computing, Год журнала: 2023, Номер 137, С. 110172 - 110172

Опубликована: Март 4, 2023

Язык: Английский

---

Next-level vegetation health index forecasting: A ConvLSTM study using MODIS Time Series

Environmental Science and Pollution Research, Год журнала: 2024, Номер 31(12), С. 18932 - 18948

Опубликована: Фев. 14, 2024

Abstract The Vegetation Health Index (VHI) is a metric used to assess the health and condition of vegetation, based on satellite-derived data. It offers comprehensive indicator stress or vigor, commonly in agriculture, ecology, environmental monitoring for forecasting changes vegetation health. Despite its advantages, there are few studies VHI as future projection, particularly using up-to-date effective machine learning methods. Hence, primary objective this study forecast values by utilizing remotely sensed images. To achieve objective, proposes employing combined Convolutional Neural Network (CNN) specific type Recurrent (RNN) called Long Short-Term Memory (LSTM), known ConvLSTM. time series images calculated Normalized Difference (NDVI) Land Surface Temperature (LST) data obtained from Moderate Resolution Imaging Spectroradiometer (MODIS) aboard Terra Aqua satellites. In addition traditional image-based calculation, suggests global minimum maximum (global scale) NDVI LST calculating VHI. results showed that ConvLSTM with 1-layer structure generally provided better forecasts than 2-layer 3-layer structures. average Root Mean Square Error (RMSE) 1-step, 2-step, 3-step ahead were 0.025, 0.026, respectively, each step representing an 8-day horizon. Moreover, proposed scale model applied structures outperformed calculation method.

Язык: Английский

---

Predicting oil palm yield using a comprehensive agronomy dataset and 17 machine learning and deep learning models

Ecological Informatics, Год журнала: 2024, Номер 81, С. 102595 - 102595

Опубликована: Апрель 17, 2024

The rising global demand for oil palm emphasizes the importance of accurate yield predictions. This predictive capability is critical effective crop management, supply chain optimization, and sustainable farming practices. However, sector faces challenges in projection, stressing a noteworthy gap application evaluation modern machine learning deep technologies. Our study addressed this by systematically evaluating 17 models predicting yield, incorporating various agronomic parameters, e.g., soil composition, climatic conditions, plant age, techniques. holistic approach enhances agriculture. Using feature selection technique maximum depth 32 1000 estimators, Extra Trees Regressor exhibited positive performance, i.e., MSE = 860.36 an R2 0.65, stands out among evaluated. findings also showed that comprehensive dataset to prediction. Hence, model have potential be robust decision-making tool agronomists farmers industry, setting stage future innovations agriculture

Язык: Английский

---

Analysis of Dieback in a Coastal Pinewood in Campania, Southern Italy, through High-Resolution Remote Sensing

Plants, Год журнала: 2024, Номер 13(2), С. 182 - 182

Опубликована: Янв. 9, 2024

For some years, the stone pine (Pinus pinea L.) forests of Domitian coast in Campania, Southern Italy, have been at risk conservation due to biological adversities. Among these, tortoise scale Toumeyella parvicornis (Cockerell) has assumed a primary role since its spread Campania began. Observation using remote sensing techniques was useful for acquiring information on health state vegetation. In this way, it possible monitor functioning forest ecosystem and identify existence critical states. To study variation spectral behavior conditions plant stress action pests, analysis multispectral data Copernicus Sentinel-2 satellite, acquired over seven years between 2016 2022, conducted forest. This method used plot values individual pixels time by processing indices Geographic Information System (GIS) tools. The use vegetation made highlight degradation suffered infestation T. parvicornis. results showed utility monitoring through high-resolution protect preserve peculiar coast.

Язык: Английский

---