Low‐Altitude UAV Imaging Accurately Quantifies Eelgrass Wasting Disease From Alaska to California

Declines in eelgrass, an important and widespread coastal habitat, are associated with wasting disease in recent outbreaks on the Pacific coast of North America. This study presents a novel method for mapping and predicting wasting disease using Unoccupied Aerial Vehicle (UAV) with low‐altitude autonomous imaging of visible bands. We conducted UAV mapping and sampling in intertidal eelgrass beds across multiple sites in Alaska, British Columbia, and California. We designed and implemented a UAV low‐altitude mapping protocol to detect disease prevalence and validated against in situ results. Our analysis revealed that green leaf area index derived from UAV imagery was a strong and significant (inverse) predictor of spatial distribution and severity of wasting disease measured on the ground, especially for regions with extensive disease infection. This study highlights a novel, efficient, and portable method to investigate seagrass disease at landscape scales across geographic regions and conditions.Alternate abstract:Plain Language SummaryDiseases of marine organisms are increasing in many regions worldwide, therefore, efficient time‐series monitoring is critical for understanding the dynamics of disease and examining its progression in time to implement management interventions. In the first study of its kind, we use high‐resolution Unoccupied Aerial Vehicle (UAV) imagery collected to detect disease at 12 sites across the North‐East Pacific coast of North America spanning 18 degrees of latitude. The low altitude UAV visible‐bands imagery achieved 1.5 cm spatial resolution, and analysis was performed at the seagrass leaf scale based on object‐oriented image analysis. Our findings suggest that drone mapping of coastal plants may substantially increase the scale of disease risk assessments in nearshore habitats and further our understanding of seagrass meadow spatial‐temporal dynamics. These can be scaled up by searching for environmental signals of the pathogen, for example, with surveillance of wastewater for signs of Covid in human populations. This application could easily apply to other areas to construct a high‐resolution monitoring network for seagrass conservation.

Exploring Transferable Techniques to Retrieve Crop Biophysical and Biochemical Variables Using Sentinel-2 Data

The current study aimed to determine the spatial transferability of eXtreme Gradient Boosting (XGBoost) models for estimating biophysical and biochemical variables (BVs), using Sentinel-2 data. The specific objectives were to: (1) assess the effect of different proportions of training samples (i.e., 25%, 50%, and 75%) available at the Target site (DT) on the spatial transferability of the XGBoost models and (2) evaluate the effect of the Source site (DS) (i.e., trained) model accuracy on the Target site (i.e., unseen) retrieval uncertainty. The results showed that the Bothaville (DS) → Harrismith (DT) Leaf Area Index (LAI) models required only fewer proportions, i.e., 25% or 50%, of the training samples to make optimal retrievals in the DT (i.e., RMSE: 0.61 m2 m−2;R2: 59%), while Harrismith (DS) →Bothaville (DT) LAI models required up to 75% of training samples in the DT to obtain optimal LAI retrievals (i.e., RMSE = 0.63 m2 m−2;R2 = 67%). In contrast, the chlorophyll content models for Bothaville (DS) → Harrismith (DT) required significant proportions of samples (i.e., 75%) from the DT to make optimal retrievals of Leaf Chlorophyll Content (LCab) (i.e., RMSE: 7.09 µg cm−2;R2: 58%) and Canopy Chlorophyll Content (CCC) (i.e., RMSE: 36.3 µg cm−2;R2: 61%), while Harrismith (DS) →Bothaville (DT) models required only 25% of the samples to achieve RMSEs of 8.16 µg cm−2 (R2: 83%) and 40.25 µg cm−2 (R2: 77%), for LCab and CCC, respectively. The results also showed that the source site model accuracy led to better transferability for LAI retrievals. In contrast, the accuracy of LCab and CCC source site models did not necessarily improve their transferability. Overall, the results elucidate the potential of transferable Machine Learning Regression Algorithms and are significant for the rapid retrieval of important crop BVs in data-scarce areas, thus facilitating spatially-explicit information for site-specific farm management.

Radiation estimation and crop growth trajectory reconstruction by novel algorithms improve MOD16 evapotranspiration predictability for global multi-site paddy rice ecosystems

• MOD16 products indicated significant underestimations in all paddy rice ET observations. • R n estimation in overcast conditions and LAI reconstruction were two key causes. • Daily R n estimations under all-sky conditions by a global cloudy index algorithm were improved by 40.6%. • Daily LAI dynamics estimated by the LTDG_PhenoS algorithm were improved by 818.7%. • Daily ET estimations were improved by 68.7%. Reliable estimations in evapotranspiration (ET) of paddy rice ecosystems by satellite products are critical because of their important roles in regional hydrological processes and climate change. However, the NASA MODIS ET products (MOD16A2) and its derivatives do not have good correlations with all global paddy rice ET observations. In this research, MOD16 model sensitivity analyses and parameter optimization strategies were conducted in order to solve the problem. Results suggested that underestimation of daily net radiation (R n) in overcast conditions and less satisfactory reconstruction of field-scale leaf area index (LAI) growth trajectory from the start date of field flooding and transplanting (FFTD) to the end of growing seasons by MODIS coarse vegetation index were identified as two major causes. A Light and Temperature-Driven Growth model and a Phenology-based LAI temporal Smoothing method fusion algorithm (LTDG_PhenoS) and an improved R n estimation method were introducted and evaluated in paddy rice fields in South Korea, Japan, China, Philippines, India, Spain, Italy, and the USA from 2002 to 2019. The LTDG_PhenoS algorithm considers Landsat and MODIS EVI observations and meteorological data as input variables and 30-m LAI daily time series as outcomes. Introducing the global cloudy index algorithm resulted in improved estimations of daily R n under all-sky conditions, with a significant decrease of root mean square error (RMSE) from 1.87 to 1.11 MJ m−2 day−1. The LTDG_PhenoS algorithm well reconstructed crop LAI growth dynamics from the FFTD to the end of rice growing seasons, with a substantial decline of RMSE from 1.49 to 0.27 m2/m−2. The FFTD estimations by the LTDG_PhenoS algorithm had an R2 of 0.97 and a small RMSE of less than 12-days. Daily ET rates estimated by novel algorithms had a substantial decline in RMSE from 2.88 to 0.90 mm day−1. [ FROM AUTHOR] Copyright of Journal of Hydrology is the property of Elsevier B.V. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full . (Copyright applies to all s.)

Remote Prediction of Oilseed Rape Yield via Gaofen-1 Images and a Crop Model

The fast and accurate prediction of crop yield at the regional scale is of great significance to food policies or trade. In this study, a new model is developed to predict the yield of oilseed rape from high-resolution remote sensing images. In order to derive this model, the ground experiment and remote sensing data analysis are carried out successively. In the ground experiment, the leaf area index (LAI) of four growing stages are measured, and a regression model is established to predict yield from ground LAI. In the remote sensing analysis, a new model is built to predict ground LAI from Gaofen-1 images where the simple ratio vegetation index at the bolting stage and the VARIgreen vegetation index at the flowering stage are used. The WOFOSTWOrld FOod STudy (WOFOST) crop model is used to generate time-series ground LAI from discontinuous ground LAI, which is calibrated coarsely with the MODerate resolution imaging spectroradiometer LAI product and finely with the ground-measured data. By combining the two conclusive formulas, an estimation model is built from Gaofen-1 images to the yield of oilseed rape. The effectiveness of the proposed model is verified in Wuxue City, Hubei Province from 2014 to 2019, with the pyramid bottleneck residual network to extract oilseed rape planting areas, the proposed model to estimate yields, and the China statistical yearbooks for comparison. The validation shows that the prediction error of the proposed algorithm is less than 5.5%, which highlights the feasibility of our method for accurate prediction of the oilseed rape yield in a large area. [ FROM AUTHOR] Copyright of Remote Sensing is the property of MDPI and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full . (Copyright applies to all s.)

Response of Summer Maize Growth and Water Use to Different Irrigation Regimes

Summer maize crop development, yield, and water use characteristics under water deficit conditions at different growth stages were investigated in this study using different irrigation regime treatments at the seedling (S), jointing (J), tasseling (T), and grain filling stages (F) in 2018 and 2019 in China. Ten different irrigation treatments were set, including three-irrigation application intervals (JTFi, STFi, SJFi, SJTi), two-irrigation applications (STi, JTi, JFi), and single-irrigation applications (Ti, Ji). These were compared to the control treatment (CK), which had sufficient irrigation provided at four intervals (SJTFi). The results showed that compared to CK, a water deficit at the seedling and jointing stages had a greater effect on plant height, whereas a water deficit at the tasseling and filling stages had a greater effect on the leaf area index, and a continuous water deficit had an effect on the stem diameter of summer maize. Limitations in terms of the growth and development of summer maize increased with less frequent irrigation. As irrigation decreased, the grain yield decreased, and the water use efficiency increased, and a water deficit at the tasseling stage had the greatest effect on the yield and water use efficiency. The JTFi treatment was the optimal irrigation regime with a yield decline, and its water consumption was reduced by 16.9% (p < 0.05) on average. However, compared to CK, the water use efficiency of the JTFi treatment increased by 17.3% (p < 0.05). Moreover, the JTFi treatment had the smallest maize yield response factor value (Ky) of 0.16, and its comprehensive score was the second highest after CK. [ FROM AUTHOR] Copyright of Agronomy is the property of MDPI and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full . (Copyright applies to all s.)

Leaf Area Index Estimation of Pergola-Trained Vineyards in Arid Regions Based on UAV RGB and Multispectral Data Using Machine Learning Methods

The leaf area index (LAI), a valuable variable for assessing vine vigor, reflects nutrient concentrations in vineyards and assists in precise management, including fertilization, improving yield, quality, and vineyard uniformity. Although some vegetation indices (VIs) have been successfully used to assess LAI variations, they are unsuitable for vineyards of different types and structures. By calibrating the light extinction coefficient of a digital photography algorithm for proximal LAI measurements, this study aimed to develop VI-LAI models for pergola-trained vineyards based on high-resolution RGB and multispectral images captured by an unmanned aerial vehicle (UAV). The models were developed by comparing five machine learning (ML) methods, and a robust ensemble model was proposed using the five models as base learners. The results showed that the ensemble model outperformed the base models. The highest R2 and lowest RMSE values that were obtained using the best combination of VIs with multispectral data were 0.899 and 0.434, respectively;those obtained using the RGB data were 0.825 and 0.547, respectively. By improving the results by feature selection, ML methods performed better with multispectral data than with RGB images, and better with higher spatial resolution data than with lower resolution data. LAI variations can be monitored efficiently and accurately for large areas of pergola-trained vineyards using this framework.