Agroforestry has large potential for carbon (C) sequestration while providing many economical, social, and ecological benefits via its diversified products. Airborne lidar is considered as the most accurate technology for mapping aboveground biomass (AGB) over landscape levels. However, little research in the past has been done to study AGB of agroforestry systems using airborne lidar data. Focusing on an agroforestry system in the Brazilian Amazon, this study first predicted plot-level AGB using fixed-effects regression models that assumed the regression coefficients to be constants. The model prediction errors were then analyzed from the perspectives of tree DBH (diameter at breast height)—height relationships and plot-level wood density, which suggested the need for stratifying agroforestry fields to improve plot-level AGB modeling. We separated teak plantations from other agroforestry types and predicted AGB using mixed-effects models that can incorporate the variation of AGB-height relationship across agroforestry types. We found that, at the plot scale, mixed-effects models led to better model prediction performance (based on leave-one-out cross-validation) than the fixed-effects models, with the coefficient of determination (R²) increasing from 0.38 to 0.64. At the landscape level, the difference between AGB densities from the two types of models was ~10% on average and up to ~30% at the pixel level. This study suggested the importance of stratification based on tree AGB allometry and the utility of mixed-effects models in modeling and mapping AGB of agroforestry systems.

Cotton plays an essential role in global human life and economic development. However, diseases such as leaf blight pose a serious threat to cotton production. This study aims to advance the existing approach by identifying cotton blight infection and classifying its severity at a higher accuracy. We selected a cotton field in Shihezi, Xinjiang in China to acquire multispectral images with an unmanned airborne vehicle (UAV); then, fifty-three 50 cm by 50 cm ground framed plots were set with defined coordinates, and a photo of its cotton canopy was taken of each and converted to the L*a*b* color space as either a training or a validation sample; finally, these two kinds of images were processed and combined to establish a cotton blight infection inversion model. Results show that the Red, Rededge, and NIR bands of multispectral UAV images were found to be most sensitive to changes in cotton leaf color caused by blight infection; NDVI and GNDVI were verified to be able to infer cotton blight infection information from the UAV images, of which the model calibration accuracy was 84%. Then, the cotton blight infection status was spatially identified with four severity levels. Finally, a cotton blight inversion model was constructed and validated with ground framed photos to be able to explain about 86% of the total variance. Evidently, multispectral UAV images coupled with ground framed cotton canopy photos can improve cotton blight infection identification accuracy and severity classification, and therefore provide a more reliable approach to effectively monitoring such cotton disease damage.

Using remote sensing data to estimate forest aboveground biomass (FAB) is of great significance for understanding the terrestrial C dynamics and making regional policies of forest management. The vegetation type-specified approach can improve the accuracy of estimating FAB, and it is difficult to obtain enough field plots for each vegetation type to fit a statistic model in practice. In this study, we tried to solve this problem by fitting a linear mixed-effects model based on the Landsat TM image and vegetation types. For developing the most parsimonious mixed-effect models, we used stepwise regressions to fit a linear regression model at the beginning and then we extended it into a linear mixed-effects model with the backward elimination method by adding and testing random effects to the explanatory variables in the linear regression model. The results showed that the linear mixed-effects model, compared with the linear regression model, has higher estimation accuracy with fewer parameters and smaller RMSE (R² = 0.63, RMSE reduced by 14.81%). The linear mixed-effects model is a feasible way to fit a species-level FAB estimation model when the field plots data are limited and uneven for different vegetation types.

Accurate and efficient crop classification is fundamental to smart agriculture. Due to the limitations of optical remote sensing in cloudy and rainy regions, time-series Synthetic Aperture Radar (SAR) data has become the primary source for crop classification, owing to its all-weather capabilities and sensitivity to crop phenology. However, most existing classification algorithms are restricted to using either temporal or spatial features alone, hindering the full exploitation of the spatial structures and temporal scattering characteristics embedded in time-series SAR data. This study proposes a CNN-BiLSTM model that combines the strengths of Convolutional Neural Networks (CNN) in spatial feature extraction with the capability of Bidirectional Long Short-Term Memory (BiLSTM) networks to capture long-term temporal dependencies, achieving deep spatiotemporal coupling of time-series SAR data. The proposed approach overcomes the limitations of traditional models that rely on single-feature utilization, constructing a more comprehensive feature representation for crop classification. A case study was conducted using nine Sentinel-1A images from the main growing season (March–June 2023) in Leizhou City, Guangdong Province. Experimental results demonstrate that the CNN-BiLSTM model significantly outperforms Support Vector Machine (SVM), CNN, and BiLSTM models in crop classification, achieving an overall accuracy of 95.8% and a Kappa coefficient of 0.94. This approach provides a novel solution for crop classification in cloudy and rainy regions and offers important applications for precise monitoring and management in smart agriculture.

The lockdowns imposed worldwide to curb the coronavirus diseases (COVID-19) spread has positive effects on the environment. However, it is unclear what fraction is caused by weather and what is related to lockdown. Here we used Global Navigation Satellite System (GNSS) height anomaly time series to quantify the spatio-temporal characteristics of lockdown-induced noise anomalies at 231 selected sites over mainland China. The results indicated that apparent declines (0.52 mm) in noise in 6 days (24–29 January 2020) after the lockdown in Wuhan resulted in drastically reduced human activities, which accounts for 71% of the total noise decrease. The lockdown effects persisted for 8 weeks and reached the maximum in the third week (6–12 February 2020) with reduced GNSS noise anomalies occurring at 81% of the GNSS sites. With the control of pandemic, increased noise anomalies occurred at more than 60% of the sites during the 9th week, which correlated well with the easing of lockdown in many cities in China. We concluded that this study provides new insights to quantifying the effects of human activities on geodetic measurements during the COVID-19 lockdown.

The pine wood nematode (PWN), one of the globally significant forest diseases, has driven the demand for precise detection methods. Recent advances in satellite remote sensing technology, particularly ultra-high-resolution optical imagery, have opened new avenues for identifying PWN-infected trees. In order to systematically evaluate the ability of ultra-high-resolution optical remote sensing and the influence of spatial and spectral resolution in detecting PWN-infected trees, this study utilized a U-Net network model to identify PWN-infected trees using three remote sensing datasets of the ultra-high-resolution multispectral imagery from Beijing 3 International Cooperative Remote Sensing Satellite (BJ3N), with a panchromatic band spatial resolution of 0.3 m and six multispectral bands at 1.2 m; the high-resolution multispectral imagery from the Beijing 3A satellite (BJ3A), with a panchromatic band resolution of 0.5 m and four multispectral bands at 2 m; and unmanned aerial vehicle (UAV) imagery with five multispectral bands at 0.07 m. Comparison of the identification results demonstrated that (1) UAV multispectral imagery with 0.07 m spatial resolution achieved the highest accuracy, with an F1 score of 89.1%. Next is the fused ultra-high-resolution BJ3N satellite imagery at 0.3 m, with an F1 score of 88.9%. In contrast, BJ3A imagery with a raw spatial resolution of 2 m performed poorly, with an F1 score of only 28%. These results underscore that finer spatial resolution in remote sensing imagery directly enhances the ability to detect subtle canopy changes indicative of PWN infestation. (2) For UAV, BJ3N, and BJ3A imagery, the identification accuracy for PWN-infected trees showed no significant differences across various band combinations at equivalent spatial resolutions. This indicates that spectral resolution plays a secondary role to spatial resolution in detecting PWN-infected trees using ultra-high-resolution optical imagery. (3) The 0.3 m BJ3N satellite imagery exhibits low false-detection and omission rates, with F1 scores comparable to higher-resolution UAV imagery. This indicates that a spatial resolution of 0.3 m is sufficient for identifying PWN-infected trees and is approaching a point of saturation in a subtropical mountain monsoon climate zone. In conclusion, ultra-high-resolution satellite remote sensing, characterized by frequent data revisit cycles, broad spatial coverage, and balanced spatial-spectral performance, provides an optimal remote sensing data source for identifying PWN-infected trees. As such, it is poised to become a cornerstone of future research and practical applications in detecting and managing PWN infestations globally.