Estimation methods of wetland carbon sink and factors influencing wetland carbon cycle: a review

Carbon Research, Journal Year: 2024, Volume and Issue: 3(1)

Published: May 20, 2024

Abstract In the global ecosystem, wetlands are vital carbon sinks, playing a crucial role in absorbing greenhouse gases such as dioxide and mitigating warming. Accurate estimation of wetland content is essential for research on sinks. However, cycle complex, sinking affected by climate, topography, water level conditions, vegetation types, soil other factors. This has caused significant challenges current studies, most focused impact individual factors often ignoring interaction between various factors, which further leads to uncertainty measurements. paper aims elucidate process cycle, summarize affecting explore interplay their influence aiming provide theoretical support study Additionally, this reviews advantages disadvantages measurement methods, proposes directions combining machine learning identifies existing difficulties measurement, offers suggestions serve reference future sink management. Graphical

Language: Английский

---

Investigation of the influence of geographical factors on soil suitability using a nonparametric controlled method of training and data analysis

E3S Web of Conferences, Journal Year: 2023, Volume and Issue: 431, P. 03005 - 03005

Published: Jan. 1, 2023

This paper analysed a dataset using selected data analysis tool. The study found that decision tree was suitable tool to analyse this set. Special attention given the of geographical factors including an assessment presence water bodies in county. showed these have significant impact on soil workability. Although model based did not absolute accuracy (14% error), it still acceptable and cheaper implement. One main advantages predict workability is their easy availability. Data other indicators can be easily used analysis. thus confirms effectiveness combination with datasets related serviceability. Despite some inaccuracy model, its relative simplicity accessibility make attractive for forecasting making area.

Language: Английский

---

Ada-XG-CatBoost: A Combined Forecasting Model for Gross Ecosystem Product (GEP) Prediction

Sustainability, Journal Year: 2024, Volume and Issue: 16(16), P. 7203 - 7203

Published: Aug. 22, 2024

The degradation of the ecosystem and loss natural capital have seriously threatened sustainable development human society economy. Currently, most research on Gross Ecosystem Product (GEP) is based statistical modeling methods, which face challenges such as high difficulty, costs, inaccurate quantitative methods. However, machine learning models are characterized by efficiency, fewer parameters, higher accuracy. Despite these advantages, their application in GEP not widespread, particularly area combined models. This paper includes both a combination model an explanatory analysis model. first to propose prediction called Ada-XGBoost-CatBoost (Ada-XG-CatBoost), integrates Extreme Gradient Boosting (XGBoost), Categorical (CatBoost) algorithms, SHapley Additive exPlanations (SHAP) approach overcomes limitations single-model evaluations aims address current issues incomplete assessments. It provides new guidance methods for enhancing value services achieving regional development. Based actual ecological data national city, preprocessing feature correlation carried out using XGBoost CatBoost AdaGrad optimization algorithm, Bayesian hyperparameter method. By selecting 11 factors that predominantly influence GEP, training selected datasets, optimizing error gradient then updated adjust weights, minimizes errors. reduces risk overfitting individual enhances predictive accuracy interpretability results indicate mean squared (MSE) Ada-XG-CatBoost reduced 65% 70% compared CatBoost, respectively. Additionally, absolute (MAE) 4.1% 42.6%, Overall, has more accurate stable performance, providing accurate, efficient, reliable reference industry.

Language: Английский

---

Vegetation dynamics in Mainland Southeast Asia: Climate and anthropogenic influences

Land Use Policy, Journal Year: 2025, Volume and Issue: 153, P. 107546 - 107546

Published: March 29, 2025

Language: Английский

---

A comparison of machine learning algorithms for predicting gross primary productivity of the Western Ghats region in India using reanalysis and satellite data

Journal of Earth System Science, Journal Year: 2025, Volume and Issue: 134(2)

Published: April 12, 2025

Language: Английский

---

Global-scale improvement of the estimation of terrestrial gross primary productivity by integrating optical and microwave remote sensing with meteorological data

Ecological Informatics, Journal Year: 2024, Volume and Issue: 83, P. 102780 - 102780

Published: Aug. 23, 2024

Photosynthesis (a key ecological process) is measured based on gross primary productivity (GPP), emphasizing the criticality of accurate GPP estimation to climate change research. The extant remote sensing-based approaches for were typically optical sensing data, neglecting potential supplementary information from microwave data. Thus, random forest algorithm, we developed a model through integration and with meteorological data (OMM-GPP). software tools used processing, modeling, analysis are standard third-party libraries Python Language. Our OMM-GPP was trained validated using (referred as "observed GPP") retrieved carbon dioxide fluxes at 137 flux towers. results indicated that by integrating more than single-source (optical or data) across eight vegetation types. performed well daily monthly scales, determination coefficients (R2) (root-mean-square errors, RMSE) 0.85 (1.52 gC m−2 d−1) 0.83 (1.49 d−1), respectively, which increased (decreased) 0.17–0.03 (0.58–0.12 0.11–0.02 (0.39–0.10 compared R2 (RMSE) values obtained variables. Further, effectively captured seasonal variations in most eight-day-scale comparison VODCA2GPP dataset revealed its enhanced performances, increasing 0.33 decreasing RMSE 0.97 d−1. Overall, can enhance accuracy, demonstrated established OMM-GPP, different

Language: Английский

---

Grazing intensity estimation in temperate typical grasslands of Inner Mongolia using machine learning models

Ecological Indicators, Journal Year: 2025, Volume and Issue: 172, P. 113318 - 113318

Published: March 1, 2025

Language: Английский

---

Impacts of climate extremes on variations in evergreen forest ecosystem carbon–water fluxes across Southern China

Global and Planetary Change, Journal Year: 2025, Volume and Issue: unknown, P. 104867 - 104867

Published: May 1, 2025

Language: Английский

---

Predicting Gross Primary Productivity under Future Climate Change for the Tibetan Plateau Based on Convolutional Neural Networks

Remote Sensing, Journal Year: 2024, Volume and Issue: 16(19), P. 3723 - 3723

Published: Oct. 7, 2024

Gross primary productivity (GPP) is vital for ecosystems and the global carbon cycle, serving as a sensitive indicator of ecosystems’ responses to climate change. However, impact future changes on GPP in Tibetan Plateau, an ecologically important climatically region, remains underexplored. This study aimed develop data-driven approach predict seasonal annual variations Plateau up year 2100 under changing climatic conditions. A convolutional neural network (CNN) was employed investigate relationships between various environmental factors, including variables, CO2 concentrations, terrain attributes. analyzed projected from Coupled Model Intercomparison Project Phase 6 (CMIP6) four scenarios: SSP1–2.6, SSP2–4.5, SSP3–7.0, SSP5–8.5. The results suggest that expected significantly increase throughout 21st century all scenarios. By 2100, reach 1011.98 Tg C, 1032.67 1044.35 1055.50 C scenarios, representing 0.36%, 4.02%, 5.55%, 5.67% relative 2021. analysis indicates spring autumn shows more pronounced growth SSP3–7.0 SSP5–8.5 scenarios due extended growing season. Furthermore, identified elevation band 3000 4500 m particularly change terms response. Significant increases would occur east Qilian Mountains upper reaches Yellow Yangtze Rivers. These findings highlight pivotal role driving dynamics this region. insights not only bridge existing knowledge gaps regarding over coming decades but also provide valuable guidance formulation adaptation strategies at ecological conservation management.

Language: Английский

---

Estimation of Daily Maize Gross Primary Productivity by Considering Specific Leaf Nitrogen and Phenology via Machine Learning Methods

Remote Sensing, Journal Year: 2024, Volume and Issue: 16(2), P. 341 - 341

Published: Jan. 15, 2024

Maize gross primary productivity (GPP) contributes the most to global cropland GPP, making it crucial accurately estimate maize GPP for carbon cycle. Previous research validated machine learning (ML) methods using remote sensing and meteorological data plant yet they disregard vegetation physiological dynamics driven by phenology. Leaf nitrogen content per unit leaf area (i.e., specific (SLN)) greatly affects photosynthesis. Its maximum allowable value correlates with a phenological factor conceptualized as normalized phenology (NMP). This study aims validate SLN NMP estimation four ML (random forest (RF), support vector (SVM), convolutional neutral network (CNN), extreme (ELM)). Inputs consist of index (NDVI), air temperature, solar radiation (SSR), NMP, SLN. Data from American flux sites (NE1, NE2, NE3 in Nebraska RO1 site Minnesota) were gathered. Using three NE effect MMP shows that accuracy notably increased after adding MMP. Among these methods, RF SVM achieved best performance Nash–Sutcliffe efficiency coefficient (NSE) = 0.9703 0.9706, root mean square error (RMSE) 1.5596 1.5509 gC·m−2·d−1, variance (CV) 0.1508 0.1470, respectively. When evaluating models at site, only CNN could effectively incorporate impact NMP. But, terms unbiased results, comprehensively enhanced Due their fixed relationship, introducing or alone might be more effective than both simultaneously, considering redundancy like ELM. supports integration leaf-level photosynthetic factors via provides reference similar research.

Language: Английский

---