Understanding the spatial distribution of dead pine trees (DPTs) in mountainous areas is vital for diseased wood cleanup and the prediction of pine wilt disease. The study induced a deep learning (DL) model of convolutional neural networks (CNNs) for the DPT detection using unmanned aerial vehicle images. Two thousand manually labelled DPT and Non-DPT samples were collected to train (80%) and optimize (20%) the CNNs of AlexNet and GoogLeNet, and 768 samples from other three areas were used to predict the categories in Jinjiang, Fujian, southeastern China. The same dataset was used to compare the classification accuracy between CNNs and the traditional template matching (TM) method. In addition, the potential factors influencing the overall accuracy of the DL model for DPT detection were evaluated. Training results of AlexNet and GoogLeNet showed an accuracy of over 93.30% and 97.38% respectively for the top-1 test after the final training iteration. For the DPT classification in the prediction areas of pure forests, the TM method showed the 56–65% of overall accuracy while CNNs showed 65–80%. CNNs-based DPT detection was more accurate than the traditional TM method. Terrains and red broadleaf trees influence the CNNs-based DPT detection from aerial visible-images in mixed artificial forest and peaks areas.

Introduction: Leaf area index (LAI) is a critical physiological and biochemical parameter that profoundly affects vegetation growth. Accurately estimating the LAI for winter wheat during jointing stage is particularly important for monitoring wheat growth status and optimizing variable fertilization decisions. Recently, unmanned aerial vehicle (UAV) data and machine/depth learning methods are widely used in crop growth parameter estimation. In traditional methods, vegetation indices (VI) and texture are usually to estimate LAI. Plant Height (PH) unlike them, contains information about the vertical structure of plants, which should be consider.Methods: Taking Xixingdian Township, Cangzhou City, Hebei Province, China as the research area in this paper, and four machine learning algorithms, namely, support vector machine(SVM), back propagation neural network (BPNN), random forest (RF), extreme gradient boosting (XGBoost), and two deep learning algorithms, namely, convolutional neural network (CNN) and long short-term memory neural network (LSTM), were applied to estimate LAI of winter wheat at jointing stage by integrating the spectral and texture features as well as the plant height information from UAV multispectral images. Initially, Digital Surface Model (DSM) and Digital Orthophoto Map (DOM) were generated. Subsequently, the PH, VI and texture features were extracted, and the texture indices (TI) was further constructed. The measured LAI on the ground were collected for the same period and calculated its Pearson correlation coefficient with PH, VI and TI to pick the feature variables with high correlation. The VI, TI, PH and fusion were considered as the independent features, and the sample set partitioning based on joint x-y distance (SPXY) method was used to divide the calibration set and validation set of samples.Results: The ability of different inputs and algorithms to estimate winter wheat LAI were evaluated. The results showed that (1) The addition of PH as a feature variable significantly improved the accuracy of the LAI estimation, indicating that wheat plant height played a vital role as a supplementary parameter for LAI inversion modeling based on traditional indices; (2) The combination of texture features, including normalized difference texture indices (NDTI), difference texture indices (DTI), and ratio texture indices (RTI), substantially improved the correlation between texture features and LAI; Furthermore, multi-feature combinations of VI, TI, and PH exhibited superior capability in estimating LAI for winter wheat; (3) Six regression algorithms have achieved high accuracy in estimating LAI, among which the XGBoost algorithm estimated winter wheat LAI with the highest overall accuracy and best results, achieving the highest R² (R^2 = 0.88), the lowest RMSE (RMSE=0.69), and an RPD greater than 2 (RPD=2.54).Discussion: This study provided compelling evidence that utilizing XGBoost and integrating spectral, texture, and plant height information extracted from UAV data can accurately monitor LAI during the jointing stage of winter wheat. The research results will provide a new perspective for accurate monitoring of crop parameters through remote sensing.

The Tibetan Plateau (TP) is one of the most sensitive regions to global climate warming, not only at the inter-annual time scale but also at the altitudinal scale. We aim to investigate the contrasting effects of temperature and precipitation on vegetation greenness at different altitudes across the TP. In this study, interannual and elevational characteristics of the Normalized Difference Vegetation Index (NDVI), temperature, and precipitation were examined during the growing season from 1982 to 2015. We compared the elevational movement rates of the isolines of NDVI, temperature, and precipitation, and the sensitivities of elevational NDVI changes to temperature and precipitation. The results show that from 1982 to 2015, the elevational variation rate of isolines for NDVI mismatched with that for temperature and precipitation. The elevational movements of NDVI isolines were mostly controlled by precipitation at elevations below 2400 m and by the temperature at elevations above 2400 m. Precipitation appears to plays a role similar to temperature, and even a more effective role than the temperature at low elevations, in controlling elevational vegetation greenness changes at both spatial and interannual scales in the TP. This study highlights the regulation of temperature and precipitation on vegetation ecosystems along elevation gradients over the whole TP under global warming conditions.

Tea is an economically important crop. Evaluating the suitability of tea can better optimize the regional layout of the tea industry and provide a scientific basis for tea planting plans, which is also conducive to the sustainable development of the tea industry in the long run. Driving force analysis can be carried out to better understand the main influencing factors of tea growth. The main purpose of this study was to evaluate the suitability of tea planting in the study area, determine the prioritization of tea industry development in this area, and provide support for the government’s planning and decision making. This study used Sentinel image data to obtain the current land use data of the study area. The results show that the accuracy of tea plantation classification based on Sentinel images reached 86%, and the total accuracy reached 92%. Then, we selected 14 factors, including climate, soil, terrain, and human-related factors, using the analytic hierarchy process and spatial analysis technology to evaluate the suitability of tea cultivation in the study area and obtain a comprehensive potential distribution map of tea cultivation. The results show that the moderately suitable area (36.81%) accounted for the largest proportion of the tea plantation suitability evaluation, followed by the generally suitable area (31.40%), the highly suitable area (16.91%), and the unsuitable area (16.23%). Among these areas, the highly suitable area is in line with the distribution of tea cultivation at the Yingde municipal level. Finally, to better analyze the contribution of each factor to the suitability of tea, the factors were quantitatively evaluated by the Geodetector model. The most important factors affecting the tea cultivation suitability evaluation were temperature (0.492), precipitation (0.367), slope (0.302), and elevation (0.255). Natural factors influence the evaluation of the suitability of tea cultivation, and the influence of human factors is relatively minor. This study provides an important scientific basis for tea yield policy formulation, tea plantation site selection, and adaptation measures.

This paper used cloud computing platform of Google Earth Engine (GEE), to access and analyze Sentinel-2, MODIS and other multi-source remote sensing data in the last 5 years to monitor the growth of crops in China, especially in Hubei Province, during the period of the rapid spread of COVID-19 (i.e., from late January to mid-March 2020), and compared with the growth over the same period under similar climate conditions in the past 4 years. We further analyzed the indirect effects of COVID-19 on crop growth. The results showed that: The area of the crops with better growth (51%) was much more than which with worse growth (22%); the crops with better and worse growth were mainly distributed in the North China Plain (the main planting areas of winter wheat in China) and the South China regions (such as Guangxi, Guangdong province), respectively. The area of the crops with similar growth occupied 27%. In Hubei Province, the area of the crops with better growth (61%) was also more than which with worse growth (27%). It was found that there was no obvious effect from COVID-19 on the overall growth of crops in China during the period from late January to mid-March 2020, and the growth of crops was much better than which during the same period in previous years. The findings in this study are helpful in evaluating the impact of the COVID-19 on China's agriculture, which are conducive to serve the relevant agricultural policy formulation and to ensure food security.

Measuring forest aboveground biomass (AGB) at local to regional scales is critical to understanding their role in regional and global carbon cycles. The Three-North Shelterbelt Forest Program (TNSFP) is the largest ecological restoration project in the world, and has been ongoing for over 40 years. In this study, we developed models to estimate the planted forest aboveground biomass (PF_AGB) for Yulin, a typical area in the project. Surface reflectances in the study area from 1978 to 2013 were obtained from Landsat series images, and integrated forest z-scores were constructed to measure afforestation and the stand age of planted forest. Normalized difference vegetation index (NDVI) was combined with stand age to develop an initial model to estimate PF_AGB. We then developed additional models that added environment variables to our initial model, including climatic factors (average temperature, total precipitation, and total sunshine duration) and a topography factor (slope). The model which combined the total precipitation and slope greatly improved the accuracy of PF_AGB estimation compared to the initial model, indicating that the environmental variables related to water distribution indirectly affected the growth of the planted forest and the resulting AGB. Afforestation in the study area occurred mainly in the early 1980s and early 21st century, and the PF_AGB in 2003 was 2.3 times than that of 1998, since the fourth term TNSFP started in 2000. The PF_AGB in 2013 was about 3.33 times of that in 2003 because many young trees matured. The leave-one-out cross-validation (LOOCV) approach showed that our estimated PF_AGB had a significant correlation with field-measured data (correlation coefficient (r) = 0.89, p < 0.001, root mean square error (RMSE) = 6.79 t/ha). Our studies provided a method to estimate long time series PF_AGB using satellite repetitive measures, particularly for arid or semi-arid areas.