Soil moisture is a key parameter for the circulation of water and energy exchange between surface and the atmosphere, playing an important role in hydrology, agriculture, and meteorology. Traditional methods for monitoring soil moisture suffer from spatial discontinuity, time-consuming processes, and high costs. Remote sensing technology enables the non-destructive and efficient retrieval of land information, allowing rapid soil moisture monitoring to schedule crop irrigation and evaluate the irrigation efficiency. Satellite data with different resolutions provide different observation scales. Evaluating the accuracy of estimating soil moisture based on open and free satellite data, as well as exploring the comprehensiveness and adaptability of different satellites for soil moisture temporal and spatial observations, are important research contents of current soil moisture monitoring. The study utilized three types of satellite data, namely GF-1, Landsat-8, and GF-4, with respective temporal and spatial resolutions of 16 m (every 4 days), 30 m (every 16 days), and 50 m (daily). The gray relational analysis (GRA) was employed to identify vegetation indices that selected sensitivity to soil moisture at varying depths (3 cm, 10 cm, and 20 cm). Then, this study employed random forest (RF), Extra Tree (ETr), and linear regression (LR) algorithms to estimate soil moisture at different depths with optical satellite data sources. The results showed that the accuracy of soil moisture estimation was different at different growth stages. The model accuracy exhibited an upward trend during the middle and late growth stages, coinciding with higher vegetation coverage; however, it demonstrated a decline in accuracy during the early and late growth stages due to either the absence or limited presence of vegetation. Among the three satellite images, the vegetation indices derived from GF-1 exhibited were more sensitive to vegetation characteristics and demonstrated superior soil moisture estimation accuracy (with R² ranging 0.129–0.928, RMSE ranging 0.017–0.078), followed by Landsat-8 (with R² ranging 0.117–0.862, RMSE ranging 0.017–0.088). The soil moisture estimation accuracy of GF-4 was the worst (with R² ranging 0.070–0.921, RMSE ranging 0.020–0.140). Thus, GF-1 is suitable for vegetated areas. In addition, the ETr model outperformed the other models in both accuracy and stability (ETr model: R² ranging from 0.117 to 0.928, RMSE ranging from 0.021 to 0.091; RF model: R² ranging from 0.225 to 0.926, RMSE ranging from 0.019 to 0.085; LR model: R² ranging from 0.048 to 0.733, RMSE ranging from 0.030 to 0.144). Utilizing GF-1 is recommended to construct the ETr model for assessing soil moisture variations in the farming land of northern China. Therefore, in cases where there are limited ground sample data, it is advisable to utilize high-spatiotemporal-resolution remote sensing data, along with machine learning algorithms such as ETr and RF, which are suitable for small samples, for soil moisture estimation.

Detecting and attributing vegetation variations in the Yellow River Basin (YRB) is vital for adjusting ecological restoration strategies to address the possible threats posed by changing environments. On the basis of the kernel normalized difference vegetation index (kNDVI) and key climate drivers (precipitation (PRE), temperature (TEM), solar radiation (SR), and potential evapotranspiration (PET)) in the basin during the period from 1982 to 2022, we utilized the multivariate statistical approach to analyze the spatiotemporal patterns of vegetation dynamics, identified the key climate variables, and discerned the respective impacts of climate change (CC) and human activities (HA) on these variations. Our analysis revealed a widespread greening trend across 93.1% of the YRB, with 83.2% exhibiting significant increases in kNDVI (p < 0.05). Conversely, 6.9% of vegetated areas displayed a browning trend, particularly concentrated in the alpine and urban areas. With the Hurst index of kNDVI exceeding 0.5 in 97.5% of vegetated areas, the YRB tends to be extensively greened in the future. Climate variability emerges as a pivotal determinant shaping diverse spatial and temporal vegetation patterns, with PRE exerting dominance in 41.9% of vegetated areas, followed by TEM (35.4%), SR (13%), and PET (9.7%). Spatially, increased PRE significantly enhanced vegetation growth in arid zones, while TEM and SR controlled vegetation variations in alpine areas and non-water-limited areas such as irrigation zones. Vegetation dynamics in the YRB were driven by a combination of CC and HA, with relative contributions of 55.8% and 44.2%, respectively, suggesting that long-term CC is the dominant force. Specifically, climate change contributed to the vegetation greening seen in the alpine region and southeastern part of the basin, and human-induced factors benefited vegetation growth on the Loess Plateau (LP) while inhibiting growth in urban and alpine pastoral areas. These findings provide critical insights that inform the formulation and adaptation of ecological conservation strategies in the basin, thereby enhancing resilience to changing environmental conditions.

The accurate prediction of cropland evapotranspiration (ET) is of utmost importance for effective irrigation and optimal water resource management. To evaluate the feasibility and accuracy of ET estimation in various climatic conditions using machine learning models, three-, six-, and nine-factor combinations (V3, V6, and V9) were examined based on the data obtained from global cropland eddy flux sites and Moderate Resolution Imaging Spectroradiometer (MODIS) remote sensing data. Four machine learning models, random forest (RF), support vector machine (SVM), extreme gradient boosting (XGB), and backpropagation neural network (BP), were used for this purpose. The input factors included daily mean air temperature (T_a), net radiation (R_n), soil heat flux (G), evaporative fraction (EF), leaf area index (LAI), photosynthetic photon flux density (PPFD), vapor pressure deficit (VPD), wind speed (U), and atmospheric pressure (P). The four machine learning models exhibited significant simulation accuracy across various climate zones, reflected by their global performance indicator (GPI) values ranging from −3.504 to 0.670 for RF, −3.522 to 1.616 for SVM, −3.704 to 0.972 for XGB, and −3.654 to 1.831 for BP. The choice of suitable models and the different input factors varied across different climatic regions. Specifically, in the temperate–continental zone (TCCZ), subtropical–Mediterranean zone (SMCZ), and temperate zone (TCZ), the models of BP_C-V9, SVM_S-V6, and SVM_T-V6 demonstrated the highest simulation accuracy, with average RMSE values of 0.259, 0.373, and 0.333 mm d⁻¹, average MAE values of 0.177, 0.263, and 0.248 mm d⁻¹, average R² values of 0.949, 0.819, and 0.917, and average NSE values of 0.926, 0.778, and 0.899, respectively. In climate zones with a lower average LAI (TCCZ), there was a strong correlation between LAI and ET, making LAI more crucial for ET predictions. Conversely, in climate zones with a higher average LAI (TCZ, SMCZ), the significance of the LAI for ET prediction was reduced. This study recognizes the impact of climate zones on ET simulations and highlights the necessity for region-specific considerations when selecting machine learning models and input factor combinations.

The Loess Plateau (LP) is a typical climate-sensitive and ecologically delicate area in China. Clarifying the vegetation–climate interaction in the LP over 40+ years, particularly pre- and post-Grain to Green Program (GTGP) implementation, is crucial for addressing potential climate threats and achieving regional ecological sustainability. Utilizing the kernel Normalized Difference Vegetation Index (kNDVI) and key climatic variables (precipitation (PRE), air temperature (TEM), and solar radiation (SR)) between 1982 and 2022, we performed an extensive examination of vegetation patterns and their reaction to changes in climate using various statistical methods. Our findings highlight a considerable and widespread greening on the LP from 1982 to 2022, evidenced by a kNDVI slope of 0.0020 yr⁻¹ (p < 0.001) and a 90.9% significantly increased greened area. The GTGP expedited this greening process, with the kNDVI slope increasing from 0.0009 yr⁻¹ to 0.0036 yr⁻¹ and the significantly greened area expanding from 39.1% to 84.0%. Over the past 40 years, the LP experienced significant warming (p < 0.001), slight humidification, and a marginal decrease in SR. Post-GTGP implementation, the warming rate decelerated, while PRE and SR growth rates slightly accelerated. Since the hurst index exceeded 0.5, most of the vegetated area of the LP is expected to be greening, warming, and humidification in the future. In the long term, 75% of the LP vegetated area significantly benefited from the increase in PRE, especially in relatively dry environments. In the LP, 61% of vegetated areas showed a positive correlation between kNDVI and TEM, while 4.9% exhibited a significant negative correlation, mainly in arid zones. SR promoted vegetation growth in 23% of the vegetated area, mostly in the eastern LP. The GTGP enhanced the sensitivity of vegetation to PRE, increasing the area corresponding to a significant positive correlation from 15.3% to 59.9%. Overall, PRE has emerged as the dominant climate driver for the vegetation dynamics of the LP, followed by TEM and SR. These insights contribute to a comprehensive understanding of the climate-impact-related vegetation response mechanisms, providing guidance for efforts toward regional sustainable ecological development amid the changing climate.

The accurate estimation of reference crop evapotranspiration (ET₀) is essential for crop water consumption modeling and agricultural water resource management. In the present study, three bionic algorithms (aquila optimizer (AO), tuna swarm optimization (TSO), and sparrow search algorithm (SSA)) were combined with an extreme learning machine (ELM) model to form three mixed models (AO-ELM, TSO-ELM, and SSA-ELM). The accuracy of the ET₀ estimates for five climate regions in China from 1970 to 2019 was evaluated using the FAO-56 Penman–Monteith (P-M) equation. The results showed that the predicted values of the three mixed models and the ELM model fitted the P-M calculated values well. R² and RMSE were 0.7654–0.9864 and 0.1271–0.7842 mm·d⁻¹, respectively, for which the prediction accuracy of the AO-ELM model was the highest. The performance of the AO-ELM combination5 (maximum temperature (T_max), minimum temperature (T_min), total solar radiation (R_s), sunshine duration (n)) was most significantly improved on the basis of the ELM model. The prediction accuracy for the stations in the plateau mountain climate (PMC) region was the best, while the prediction accuracy for the stations in the tropical monsoon climate region (TPMC) was the worst. In addition to the wind speed (U₂) in the temperate continental climate region (TCC)—which was the largest variable affecting ET₀—n, R_a, and total solar radiation (R_s) in the other climate regions were more important than relative humidity (RH) and wind speed (U₂) in predicting ET₀. Therefore, AO-ELM4 was selected for the TCC region (with T_max, T_min, R_s, and U₂ as inputs) and AO-ELM5 (with T_max, T_min, R_s, and n as inputs) was selected for the TMC, PMC, SMC, and TPMC regions when determining the best model for each climate region with limited meteorological data.

The light-use efficiency-based gross primary productivity (LUE-GPP) model is widely utilized for simulating terrestrial ecosystem carbon exchanges owing to its perceived simplicity and reliability. Variations in cloud cover and aerosol concentrations can affect ecosystem LUE, thereby influencing the performance of the LUE-GPP model, particularly in humid regions. In this study, the performance of six big-leaf LUE-GPP models and one two-leaf LUE-GPP model were evaluated in a humid agroforestry ecosystem from 2018–2020. All big-leaf LUE-GPP models yielded GPP values consistent with that derived from the eddy covariance system (GPP_EC), with R² ranging from 0.66–0.73 and RMSE ranging from 1.81–3.04 g C m⁻² d⁻¹. Differences in model performance were attributed to the differences in the quantification of temperature (T_s) and moisture constraints (W_s) and their combination forms in the models. The T_s and W_s algorithms in the eddy covariance-light-use efficiency (EF-LUE) model well characterized the environmental constraints on LUE. Simulation accuracy under the common limitation of T_s and W_s (T_s × W_s) was higher than the maximum limitation of T_s or W_s (Min (T_s, W_s)), and the combination of the T_s algorithm in the Carnegie–Ames–Stanford Approach (CASA) and the W_s algorithm in the EF-LUE model was optimized in combination forms, thereby constraining LUE for GPP estimates (GPP_BLO, R² = 0.76). Various big-leaf LUE-GPP models overestimated or underestimated GPP on sunny or cloudy days, respectively, while the two-leaf LUE-GPP model, which considered the transmission of diffuse radiation and the difference in photosynthetic capacity of canopy leaves, performed well (R² = 0.72, p < 0.01). Nevertheless, the underestimation/overestimation for shaded/sunlit leaves remained under different weather conditions. Then, the clearness index (K_t) was introduced to calculate the dynamic LUE in the big-leaf and two-leaf LUE-GPP models in the form of exponential or power functions, resulting in consistent performance even in different weather conditions and an overall higher simulation accuracy. This study confirmed the potential applicability of different LUE-GPP models and emphasized the importance of dynamic LUE on model performance.