Estimating hourly air temperature in an Amazon-Cerrado transitional forest in Brazil using machine learning regression models DOI Creative Commons
Daniela Maionchi, J. S. Gonçalves,

Fábio A. Balista

et al.

Research Square (Research Square), Journal Year: 2023, Volume and Issue: unknown

Published: Oct. 12, 2023

Abstract Air temperature holds significant importance in microclimate and environmental health studies, playing a crucial role weather regulation. There is need to develop reliable model capable of accurately capturing air variations. In this study, we focused on the Amazon-Cerrado transitional forest, constructing robust predictive for hourly fluctuations. This situated approximately 50 km northwest Sinop, Mato Grosso, Brazil, area, making it important investigate its climatic behavior ecosystems. We estimated using machine learning techniques such as Random Forest, Gradient Boosting, Multilayer Perceptron, Support Vector Regressor, aiming evaluate most effective models based relevant metrics. Performance assessments were conducted during both dry rainy seasons verify their adaptability. The top-performing Forest demonstrated Willmott Spearman indexes above 0.97. relative humidity, solar radiation, volumetric soil water content identified features, evaluated with 0.95 dimensionality reduction. These results underscore efficacy estimating temperature.

Language: Английский

Satellite remote sensing of vegetation phenology: Progress, challenges, and opportunities DOI
Zheng Gong, Wenyan Ge, Jiaqi Guo

et al.

ISPRS Journal of Photogrammetry and Remote Sensing, Journal Year: 2024, Volume and Issue: 217, P. 149 - 164

Published: Aug. 29, 2024

Language: Английский

Citations

28
A Systematic Review of Applications of Machine Learning Techniques for Wildfire Management Decision Support DOI Creative Commons
Karol Bot, José G. Borges

Inventions, Journal Year: 2022, Volume and Issue: 7(1), P. 15 - 15

Published: Jan. 21, 2022

Wildfires threaten and kill people, destroy urban rural property, degrade air quality, ravage forest ecosystems, contribute to global warming. Wildfire management decision support models are thus important for avoiding or mitigating the effects of these events. In this context, paper aims at providing a review recent applications machine learning methods wildfire support. The emphasis is on summary with classification according case study type, method, location, performance metrics. considers documents published in last four years, using sample 135 (review articles research articles). It concluded that adoption may enhancing different fire phases.

Language: Английский

Citations

69
Artificial neural networks for assessing forest fire susceptibility in Türkiye DOI
Omer Kantarcioglu, Sultan Kocaman, Konrad Schindler

et al.

Ecological Informatics, Journal Year: 2023, Volume and Issue: 75, P. 102034 - 102034

Published: Feb. 26, 2023

Language: Английский

Citations

21
Spatiotemporal patterns of remotely sensed phenology and their response to climate change and topography in subtropical bamboo forests during 2001-2017: a case study in Zhejiang Province, China DOI Creative Commons
Xuejian Li, Huaqiang Du, Guomo Zhou

et al.

GIScience & Remote Sensing, Journal Year: 2023, Volume and Issue: 60(1)

Published: Jan. 3, 2023

Vegetation phenology has long been adapted to environmental change and is highly sensitive climate change. Shifts in also affect feedbacks of vegetation factors such as topography by influencing spatiotemporal fluctuations productivity, carbon fixation, the water cycle. However, there are limited studies which explores combined effects terrain on phenology. Bamboo forests exhibit outstanding phenological phenomena play an important role maintaining global balance Therefore, interaction mechanisms bamboo forest were analyzed Zhejiang Province, China during 2001–2017. The partial least squares path model was applied clarify interplay between impacts under land cover/use results revealed that average start date growing season (SOS) significantly advanced 0.81 days annually, end (EOS) delayed 0.27 length (LOS) increased 1.08 annually. There obvious spatial differences correlation coefficients metrics. Although SOS, EOS LOS affected different climatic factors, precipitation dominant factor. Due sensitivity SOS precipitation, a 100 mm increase regional annual would cause advance 0.18 be 0.12 days. Regarding affecting conditions, clear influences altitudes, slopes aspect gradients This study further showed topographic mainly interannual variations metrics precipitation. clarified pattern interactive vegetative this information crucial assessing impact sequestration potential forests.

Language: Английский

Citations

13
A comparison of physics‐based, data‐driven, and hybrid modeling approaches for rice phenology prediction DOI
Jin Yu, Yifan Zhao, Guoqing Lei

et al.

Agronomy Journal, Journal Year: 2025, Volume and Issue: 117(1)

Published: Jan. 1, 2025

Abstract Accurate prediction of paddy rice ( Oryza sativa L.) phenology is necessary for informing field management and improving yield. There exist different ways, including physics‐based, data‐driven, hybrid approaches, to make prediction. However, few studies have investigated the performance above three modeling approaches. This study compared a physics‐based model (ORYZA), data‐driven (using distributed random forest [DRF] technique), (an integration ORYZA DRF‐based development rate parameter estimates) panicle initiation flowering date The feature importance analysis method was introduced quantify relative input variables results showed following: (1) Rice genotypes cultivation patterns resulted in poor prediction, whose root mean square error (RMSE) ranged from 6.01 8.12 days, coefficient determination R 2 ) 0.06 0.24. (2) model, RMSE 3.11 3.66 improved but still underperformed 2.44 2.57 days. worse might be attributed accuracy parameter, juvenile phase, where absolute percentage 0.286. (3) Satellite‐based vegetation indices, leaf area index, evapotranspiration played an important role determining predictive capacity DRF technique parameters phenology. Overall, we suggested using models accurate

Language: Английский

Citations

0
Heterogeneous land surface phenology challenges the comparison among PlanetScope, HLS, and VIIRS detections in semi-arid rangelands DOI
Yuxia Liu, Xiaoyang Zhang, Khuong H. Tran

et al.

Agricultural and Forest Meteorology, Journal Year: 2025, Volume and Issue: 366, P. 110497 - 110497

Published: March 11, 2025

Language: Английский

Citations

0
Dynamic UAV data fusion and deep learning for improved maize phenological-stage tracking DOI Creative Commons

Ziheng Feng,

Jiliang Zhao,

Liunan Suo

et al.

The Crop Journal, Journal Year: 2025, Volume and Issue: unknown

Published: April 1, 2025

Language: Английский

Citations

0
Wildfire Prediction Model Based on Spatial and Temporal Characteristics: A Case Study of a Wildfire in Portugal’s Montesinho Natural Park DOI Open Access
Hao Dong, Han Wu, Pengfei Sun

et al.

Sustainability, Journal Year: 2022, Volume and Issue: 14(16), P. 10107 - 10107

Published: Aug. 15, 2022

Wildfires influence the global carbon cycle, and regularity of wildfires is mostly determined by elements such as meteorological conditions, combustible material states, human activities. The time series spatial dispersion have been studied some scholars. Wildfire samples were acquired in a monthly for Montesinho Natural Park historical fire site dataset (January 2000 to December 2003), which can be used assess possible effects geographical temporal variations on forest fires. Based above dataset, dynamic wildfire distribution thresholds examined using K-means++ clustering technique each subgroup, data categorized flammable or non-flammable depending thresholds. A five-fold hierarchical cross-validation strategy was train four machine learning models: extreme gradient boosting (XGBoost), random (RF), support vector (SVM), decision tree (DT). Finally, explore performance those we mentioned, accuracy (ACC), F1 score (F1), values area under curve (AUC) receiver operating characteristics (ROCs). results depicted that XGBoost model works best evaluation three metrics (ACC = 0.8132, 0.7862, AUC 0.8052). significantly improved when compared approach classifying burned size 72.3%), demonstrating spatiotemporal heterogeneity has broad occurrence. law connection could aid prediction management disasters.

Language: Английский

Citations

15
Cropland expansion delays vegetation spring phenology according to satellite and in-situ observations DOI Open Access
Guosong Zhao, Jinwei Dong, Jilin Yang

et al.

Agriculture Ecosystems & Environment, Journal Year: 2023, Volume and Issue: 356, P. 108651 - 108651

Published: July 3, 2023

Language: Английский

Citations

8
Exploring the role of biotic factors in regulating the spatial variability in land surface phenology across four temperate forest sites DOI
Yingyi Zhao, Zhihui Wang, Zhengbing Yan

et al.

New Phytologist, Journal Year: 2024, Volume and Issue: 242(5), P. 1965 - 1980

Published: April 4, 2024

Summary Land surface phenology (LSP), the characterization of plant with satellite data, is essential for understanding effects climate change on ecosystem functions. Considerable LSP variation observed within local landscapes, and role biotic factors in regulating such remains underexplored. In this study, we selected four National Ecological Observatory Network terrestrial sites minor topographic relief to investigate how regulate intra‐site variability. We utilized functional type (PFT) maps, traits, data assess explanatory power start end season (SOS EOS) Our results indicate that PFTs alone explain only 0.8–23.4% SOS EOS variation, whereas including traits significantly improves power, cross‐validation correlations ranging from 0.50 0.85. While exhibited diverse across different sites, related competitive ability productivity were important explaining both at these sites. These findings reveal plants exhibit phenological responses comparable environmental conditions, contribute variability, highlighting importance intrinsic properties phenology.

Language: Английский

Citations

2