Improving the accuracy of cotton seedling emergence rate estimation by fusing UAV-based multispectral vegetation indices

Timely and accurate estimation of cotton seedling emergence rate is of great significance to cotton production. This study explored the feasibility of drone-based remote sensing in monitoring cotton seedling emergence. The visible and multispectral images of cotton seedlings with 2 - 4 leaves in 30 plots were synchronously obtained by drones. The acquired images included cotton seedlings, bare soil, mulching films, and PE drip tapes. After constructing 17 visible VIs and 14 multispectral VIs, three strategies were used to separate cotton seedlings from the images: (1) Otsu's thresholding was performed on each vegetation index (VI); (2) Key VIs were extracted based on results of (1), and the Otsu-intersection method and three machine learning methods were used to classify cotton seedlings, bare soil, mulching films, and PE drip tapes in the images; (3) Machine learning models were constructed using all VIs and validated. Finally, the models constructed based on two modeling strategies [Otsu-intersection (OI) and machine learning (Support Vector Machine (SVM), Random Forest (RF), and K-nearest neighbor (KNN)] showed a higher accuracy. Therefore, these models were selected to estimate cotton seedling emergence rate, and the estimates were compared with the manually measured emergence rate. The results showed that multispectral VIs, especially NDVI, RVI, SAVI, EVI2, OSAVI, and MCARI, had higher crop seedling extraction accuracy than visible VIs. After fusing all VIs or key VIs extracted based on Otsu's thresholding, the binary image purity was greatly improved. Among the fusion methods, the Key VIs-OI and All VIs-KNN methods yielded less noises and small errors, with a RMSE (root mean squared error) as low as 2.69% and a MAE (mean absolute error) as low as 2.15%. Therefore, fusing multiple VIs can increase crop image segmentation accuracy. This study provides a new method for rapidly monitoring crop seedling emergence rate in the field, which is of great significance for the development of modern agriculture.

Combining machine learning and remote sensing-integrated crop modeling for rice and soybean crop simulation

Machine learning (ML) techniques offer a promising avenue for improving the integration of remote sensing data into mathematical crop models, thereby enhancing crop growth prediction accuracy. A critical variable for this integration is the leaf area index (LAI), which can be accurately assessed using proximal or remote sensing data based on plant canopies. This study aimed to (1) develop a machine learning-based method for estimating the LAI in rice and soybean crops using proximal sensing data and (2) evaluate the performance of a Remote Sensing-Integrated Crop Model (RSCM) when integrated with the ML algorithms. To achieve these objectives, we analyzed rice and soybean datasets to identify the most effective ML algorithms for modeling the relationship between LAI and vegetation indices derived from canopy reflectance measurements. Our analyses employed a variety of ML regression models, including ridge, lasso, support vector machine, random forest, and extra trees. Among these, the extra trees regression model demonstrated the best performance, achieving test scores of 0.86 and 0.89 for rice and soybean crops, respectively. This model closely replicated observed LAI values under different nitrogen treatments, achieving Nash-Sutcliffe efficiencies of 0.93 for rice and 0.97 for soybean. Our findings show that incorporating ML techniques into RSCM effectively captures seasonal LAI variations across diverse field management practices, offering significant potential for improving crop growth and productivity monitoring.

Estimation of wheat crop production using multispectral information fusion

BACKGROUND

The present work estimates the area and corresponding wheat crop production in the study area, which comprises the Etah region of Uttar Pradesh, India. For this purpose, multispectral images of multiple sensors, namely Sentinel-2, Landsat-8 and Landsat-9 during the preharvest period, i.e. March for the years 2021 and 2022, were used. A multispectral information fusion approach was proposed, involving image classification as well as vegetation index-based information extraction. For imposing information fusion, appropriate image bands were identified with the help of separability analysis followed by land cover classification for wheat crop class extraction. Support vector machine (SVM), artificial neural network (ANN) and maximum likelihood (ML) were used for classification, whereas normalized difference vegetation index (NDVI) and fractional vegetation cover (FVC) were used for index-based crop area extraction.

RESULTS

A maximum accuracy of 98.34% was achieved for Sentinel-2 data using ANN, whereas a minimum accuracy of 80.21% was achieved for Landsat-9 using the ML classifier. The estimated area for Sentinel-2 data for the year 2021 was 260 540 ha using ANN and 203 240 ha using ML, which is close to the reference data, i.e. 238 600 ha. SVM also showed good performance and calculated least error in estimated crop area for the year 2022 on Sentinel-2 data. It calculated 8 408 490 tons of wheat for the same year.

CONCLUSION

The proposed method utilizes a single image per year for extraction of information supported by the ground truth data, which makes it a novel approach to information extraction for crop production monitoring. © 2023 Society of Chemical Industry.

Remote Sensing Evaluation Drone Herbicide Application Effectiveness for Controlling Echinochloa spp. in Rice Crop in Valencia (Spain)

Rice (Oryza sativa L.) is a staple cereal in the diet of more than half of the world's population. Within the European Union, Spain is a leader in rice production due to its climate and tradition, accounting for 26% of total EU production in 2020. The Valencian rice area covers around 15,000 hectares and is strongly influenced by biotic and abiotic factors. An important biotic factor affecting rice production is weeds, which compete with rice for sunlight, water and nutrients. The dominant weed in Spain is Echinochloa spp., although wild rice is becoming increasingly important. Rice cultivation in Valencia takes place in the area of L'Albufera de Valencia, which is a natural park, i.e., a special protection area. In this natural area, the use of phytosanitary products is limited, so it is necessary to use the minimum amount possible. Therefore, the objective of this work is to evaluate the possibility of using remote sensing effectively to determine the effectiveness of the application of the herbicide cyhalofop-butyl by drone for the control of Echinochloa spp. in rice crops in Valencia. The results will be compared with those obtained by using sterilisation machines (electric backpack sprayers) to apply the herbicide. To evaluate the effectiveness of the application, the reflectance obtained by the satellite sensors in the red and near infrared (NIR) wavelengths, as well as the normalised difference vegetation index (NDVI), were used. The remote sensing results were analysed and complemented by the number of rice plants and weeds per area, plant dry weight, leaf area, BBCH phenological state, SPAD index values, chlorophyll content and relative growth rate. Remote sensing is validated as an effective tool for determining the efficacy of an herbicide in controlling weeds applied by both the drone and the electric backpack sprayer. The weeds slowed down their development after the treatment. Depending on the phenological state of the crop and the active ingredient of the herbicide, these results are applicable to other areas with different climatic and environmental conditions.

Forest Disturbance Monitoring Using Cloud-Based Sentinel-2 Satellite Imagery and Machine Learning

Forest damage has become more frequent in Hungary in the last decades, and remote sensing offers a powerful tool for monitoring them rapidly and cost-effectively. A combined approach was developed to utilise high-resolution ESA Sentinel-2 satellite imagery and Google Earth Engine cloud computing and field-based forest inventory data. Maps and charts were derived from vegetation indices (NDVI and Z∙NDVI) of satellite images to detect forest disturbances in the Hungarian study site for the period of 2017-2020. The NDVI maps were classified to reveal forest disturbances, and the cloud-based method successfully showed drought and frost damage in the oak-dominated Nagyerdő forest of Debrecen. Differences in the reactions to damage between tree species were visible on the index maps; therefore, a random forest machine learning classifier was applied to show the spatial distribution of dominant species. An accuracy assessment was accomplished with confusion matrices that compared classified index maps to field-surveyed data, demonstrating 99.1% producer, 71% user, and 71% total accuracies for forest damage and 81.9% for tree species. Based on the results of this study and the resilience of Google Earth Engine, the presented method has the potential to be extended to monitor all of Hungary in a faster, more accurate way using systematically collected field-data, the latest satellite imagery, and artificial intelligence.

Genetic Dissection of Seasonal Changes in a Greening Plant Based on Time-Series Multispectral Imaging

Good appearance throughout the year is important for perennial ornamental plants used for rooftop greenery. However, the methods for evaluating appearance throughout the year, such as plant color and growth activity, are not well understood. In this study, evergreen and winter-dormant parents of Phedimus takesimensis and 94 F1 plants were used for multispectral imaging. We took 16 multispectral image measurements from March 2019 to April 2020 and used them to calculate 15 vegetation indices and the area of plant cover. QTL analysis was also performed. Traits such as the area of plant cover and vegetation indices related to biomass were high during spring and summer (growth period), whereas vegetation indices related to anthocyanins were high in winter (dormancy period). According to the PCA, changes in the intensity of light reflected from the plants at different wavelengths over the course of a year were consistent with the changes in plant color and growth activity. Seven QTLs were found to be associated with major seasonal growth changes. This approach, which monitors not only at a single point in time but also over time, can reveal morphological changes during growth, senescence, and dormancy throughout the year.

Estimating potassium in potato plants based on multispectral images acquired from unmanned aerial vehicles

Plant potassium content (PKC) is a crucial indicator of crop potassium nutrient status and is vital in making informed fertilization decisions in the field. This study aims to enhance the accuracy of PKC estimation during key potato growth stages by using vegetation indices (VIs) and spatial structure features derived from UAV-based multispectral sensors. Specifically, the fraction of vegetation coverage (FVC), gray-level co-occurrence matrix texture, and multispectral VIs were extracted from multispectral images acquired at the potato tuber formation, tuber growth, and starch accumulation stages. Linear regression and stepwise multiple linear regression analyses were conducted to investigate how VIs, both individually and in combination with spatial structure features, affect potato PKC estimation. The findings lead to the following conclusions: (1) Estimating potato PKC using multispectral VIs is feasible but necessitates further enhancements in accuracy. (2) Augmenting VIs with either the FVC or texture features makes potato PKC estimation more accurate than when using single VIs. (3) Finally, integrating VIs with both the FVC and texture features improves the accuracy of potato PKC estimation, resulting in notable R 2 values of 0.63, 0.84, and 0.80 for the three fertility periods, respectively, with corresponding root mean square errors of 0.44%, 0.29%, and 0.25%. Overall, these results highlight the potential of integrating canopy spectral information and spatial-structure information obtained from multispectral sensors mounted on unmanned aerial vehicles for monitoring crop growth and assessing potassium nutrient status. These findings thus have significant implications for agricultural management.

Effect Analysis of Hydrogen Peroxide Using Hyperspectral Reflectance in Sorghum [Sorghum bicolor (L.) Moench] under Drought Stress

Due to global climate change, adverse environments like drought in agricultural production are occurring frequently, increasing the need for research to ensure stable crop production. This study was conducted to determine the effect of artificial hydrogen peroxide treatment on sorghum growth to induce stress resistance in drought conditions. Hyperspectral analysis was performed to rapidly find out the effects of drought and hydrogen peroxide treatment to estimate the physiological parameters of plants related to drought and calculate the vegetation indices through PLS analysis based on hyperspectral data. The partial least squares (PLS) analysis collected chlorophyll fluorescence variables, photosynthetic parameters, leaf water potential, and hyperspectral reflectance during the stem elongation and booting stage. To find out the effect of hydrogen peroxide treatment in sorghum plants grown under 90% and 60% of field capacity in greenhouses, growth and hyperspectral reflectance were measured on the 10th and 20th days after foliar application of H2O2 at 30 mM from 1st to 5th leaf stage. The PLS analysis shows that the maximum variable fluorescence of the dark-adapted leaves was the most predictable model with R2 = 0.76, and the estimation model suitability gradually increased with O (R2 = 0.51), J (R2 = 0.73), and P (R2 = 0.75) among OJIP parameters of chlorophyll fluorescence analysis. However, the estimation suitability of predictions for moisture-related traits, vapor pressure deficit (VPD, R2 = 0.18), and leaf water potential (R2 = 0.15) using hyperspectral data was low. The hyperspectral reflectance was 10% higher at 20 days after treatment (DAT) and 3% at 20 DAT than the non-treatment in the far red and infra-red light regions under drought conditions. Vogelmann red edge index (VOG REI) 1, chlorophyll index red edge (CIR), and red-edge normalized difference vegetation index (RE-NDVI) efficiently reflected moisture stress among the vegetation indices. Photochemical reflectance index (PRI) can be used as an indicator for early diagnosis of drought stress because hydrogen peroxide treatment showed higher values than untreated in the early stages of drought damage.

Abundance considerations for modeling yield of rapeseed at the flowering stage

Introduction

To stabilize the edible oil market, it is necessary to determine the oil yield in advance, so the accurate and fast technology of estimating rapeseed yield is of great significance in agricultural production activities. Due to the long flowering time of rapeseed and the characteristics of petal color that are obviously different from other crops, the flowering period can be carefully considered in crop classification and yield estimation.

Methods

A field experiment was conducted to obtain the unmanned aerial vehicle (UAV) multispectral images. Field measurements consisted of the reflectance of flowers, leaves, and soils at the flowering stage and rapeseed yield at physiological maturity. Moreover, GF-1 and Sentinel-2 satellite images were collected to compare the applicability of yield estimation methods. The abundance of different organs of rapeseed was extracted by the spectral mixture analysis (SMA) technology, which was multiplied by vegetation indices (VIs) respectively to estimate the yield.

Results

For the UAV-scale, the product of VIs and leaf abundance (AbdLF) was closely related to rapeseed yield, which was better than the VIs models for yield estimation, with the coefficient of determination (R2) above 0.78. The yield estimation models of the product of normalized difference yellowness index (NDYI), enhanced vegetation index (EVI) and AbdLF had the highest accuracy, with the coefficients of variation (CVs) below 10%. For the satellite scale, most of the estimation models of the product of VIs and rapeseed AbdLF were also improved compared with the VIs models. The yield estimation models of the product of AbdLF and renormalized difference VI (RDVI) and EVI (RDVI×AbdLF and EVI×AbdLF) had the steady improvement, with CVs below 13.1%. Furthermore, the yield estimation models of the product of AbdLF and normalized difference VI (NDVI), visible atmospherically resistant index (VARI), RDVI, and EVI had consistent performance at both UAV and satellite scales.

Discussion

The results showed that considering SMA could improve the limitation of using only VIs to retrieve rapeseed yield at the flowering stage. Our results indicate that the abundance of rapeseed leaves can be a potential indicator of yield prediction during the flowering stage.

Corrigendum: Retrieving rice (Oryza sativa L.) net photosynthetic rate from UAV multispectral images based on machine learning methods

[This corrects the article DOI: 10.3389/fpls.2022.1088499.].

From remotely-sensed solar-induced chlorophyll fluorescence to ecosystem structure, function, and service: Part II-Harnessing data

Although our observing capabilities of solar-induced chlorophyll fluorescence (SIF) have been growing rapidly, the quality and consistency of SIF datasets are still in an active stage of research and development. As a result, there are considerable inconsistencies among diverse SIF datasets at all scales and the widespread applications of them have led to contradictory findings. The present review is the second of the two companion reviews, and data oriented. It aims to (1) synthesize the variety, scale, and uncertainty of existing SIF datasets, (2) synthesize the diverse applications in the sector of ecology, agriculture, hydrology, climate, and socioeconomics, and (3) clarify how such data inconsistency superimposed with the theoretical complexities laid out in (Sun et al., 2023) may impact process interpretation of various applications and contribute to inconsistent findings. We emphasize that accurate interpretation of the functional relationships between SIF and other ecological indicators is contingent upon complete understanding of SIF data quality and uncertainty. Biases and uncertainties in SIF observations can significantly confound interpretation of their relationships and how such relationships respond to environmental variations. Built upon our syntheses, we summarize existing gaps and uncertainties in current SIF observations. Further, we offer our perspectives on innovations needed to help improve informing ecosystem structure, function, and service under climate change, including enhancing in-situ SIF observing capability especially in "data desert" regions, improving cross-instrument data standardization and network coordination, and advancing applications by fully harnessing theory and data.

An Assessment of Relations between Vegetation Green FPAR and Vegetation Indices through a Radiative Transfer Model

The fraction of absorbed photosynthetically active radiation (FPAR) is widely used in remote sensing-based production models to estimate gross or net primary production. The forest canopy is composed primarily of photosynthetically active vegetation (PAV, green leaves) and non-photosynthetic vegetation (NPV e.g., branches), which absorb PAR but only the PAR absorbed by PAV is used for photosynthesis. Green FPAR (the fraction of PAR absorbed by PAV) is essential for the accurate estimation of GPP. In this study, the scattering by arbitrary inclined leaves (SAIL) model was reconfigured to partition the PAR absorbed by forest canopies. The characteristics of green FPAR and its relationships with spectral vegetation indices (NDVI, EVI, EVI2, and SAVI) were analyzed. The results showed that green FPAR varied with the canopy structure. In the forests with high coverage, the green FPAR was close to the total FPAR, while in the open forests, the green FPAR was far smaller than the total FPAR. Plant area index had more important impacts on the green FPAR than the proportion of PAV and optical properties of PAV. The significant relationships were found between spectral vegetation indices and the green FPAR, but EVI was more suitable to describe the variation of canopy green FPAR.

Development of a Quick-Install Rapid Phenotyping System

In recent years, there has been a growing need for accessible High-Throughput Plant Phenotyping (HTPP) platforms that can take measurements of plant traits in open fields. This paper presents a phenotyping system designed to address this issue by combining ultrasonic and multispectral sensing of the crop canopy with other diverse measurements under varying environmental conditions. The system demonstrates a throughput increase by a factor of 50 when compared to a manual setup, allowing for efficient mapping of crop status across a field with crops grown in rows of any spacing. Tests presented in this paper illustrate the type of experimentation that can be performed with the platform, emphasizing the output from each sensor. The system integration, versatility, and ergonomics are the most significant contributions. The presented system can be used for studying plant responses to different treatments and/or stresses under diverse farming practices in virtually any field environment. It was shown that crop height and several vegetation indices, most of them common indicators of plant physiological status, can be easily paired with corresponding environmental conditions to facilitate data analysis at the fine spatial scale.

Inversion models of aboveground grassland biomass in Xinjiang based on multisource data

Grassland biomass monitoring is essential for assessing grassland health and carbon cycling. However, monitoring grassland biomass in drylands based on satellite remote sensing is challenging.Statistical regression models and machine learning have been used for the construction of grassland biomass models, but the predictive power for different grassland types is unclear. Additionally, the selection of the most appropriate variables to construct a biomass inversion model for different grassland types must be explored. Therefore,1201 ground-truthed data points collected from 2014-2021,including 15 Moderate Resolution Imaging Spectroradiometer (MODIS) vegetation indices,geographic location and topographic data,and meteorological factors and vegetation biophysical indicators were screened for key variables using principal component analysis (PCA). The accuracy of multiple linear regression models, exponential regression models, power function models, support vector machine (SVM) models, random forest (RF) models, and neural network models was evaluated for the inversion of three types of grassland biomass. The results were as follows: (1) The biomass inversion accuracy of single vegetation indices was low, and the optimal vegetation indices were the soil-adjusted vegetation index (SAVI) (R2 = 0.255), normalized difference vegetation index (NDVI) (R2 = 0.372) and optimized soil-adjusted vegetation index (OSAVI) (R2 = 0.285). (2)Grassland above-ground biomass (AGB) was affected by various factors such as geographic location,topography, and meteorological factors, and the inverse models using a single environmental variable had large errors. (3) The main variables used to model biomass in the three types of grasslands were different. SAVI, aspect, slope, and precipitation (Prec.) were selected for desert grasslands; NDVI,shortwave infrared 2 (SWI2), longitude, mean temperature, and annual precipitation were selected for steppe;and OSAVI, phytochrome ratio (PPR), longitude, precipitation, and temperature were selected for meadows. (4) The non-parametric meadow biomass model was superior to the statistical regression model. (5) The RF model was the best model for the inversion of grassland biomass in Xinjiang, and this model had the highest accuracy for grassland biomass inversion (R2 = 0.656, root mean square error (RMSE) = 815.6 kg/ha),followed by meadow (R2 = 0.610, RMSE = 547.9 kg/ha) and desert grassland (R2 = 0.441, RMSE = 353.6 kg/ha).

Hyperspectral remote sensing for tobacco quality estimation, yield prediction, and stress detection: A review of applications and methods

Tobacco is an important economic crop and the main raw material of cigarette products. Nowadays, with the increasing consumer demand for high-quality cigarettes, the requirements for their main raw materials are also varying. In general, tobacco quality is primarily determined by the exterior quality, inherent quality, chemical compositions, and physical properties. All these aspects are formed during the growing season and are vulnerable to many environmental factors, such as climate, geography, irrigation, fertilization, diseases and pests, etc. Therefore, there is a great demand for tobacco growth monitoring and near real-time quality evaluation. Herein, hyperspectral remote sensing (HRS) is increasingly being considered as a cost-effective alternative to traditional destructive field sampling methods and laboratory trials to determine various agronomic parameters of tobacco with the assistance of diverse hyperspectral vegetation indices and machine learning algorithms. In light of this, we conduct a comprehensive review of the HRS applications in tobacco production management. In this review, we briefly sketch the principles of HRS and commonly used data acquisition system platforms. We detail the specific applications and methodologies for tobacco quality estimation, yield prediction, and stress detection. Finally, we discuss the major challenges and future opportunities for potential application prospects. We hope that this review could provide interested researchers, practitioners, or readers with a basic understanding of current HRS applications in tobacco production management, and give some guidelines for practical works.

Multi-Annual Evaluation of Time Series of Sentinel-1 Interferometric Coherence as a Tool for Crop Monitoring

Interferometric coherence from SAR data is a tool used in a variety of Earth observation applications. In the context of crop monitoring, vegetation indices are commonly used to describe crop dynamics. The most frequently used vegetation indices based on radar data are constructed using the backscattered intensity at different polarimetric channels. As coherence is sensitive to the changes in the scene caused by vegetation and its evolution, it may potentially be used as an alternative tool in this context. The objective of this work is to evaluate the potential of using Sentinel-1 interferometric coherence for this purpose. The study area is an agricultural region in Sevilla, Spain, mainly covered by 18 different crops. Time series of different backscatter-based radar vegetation indices and the coherence amplitude for both VV and VH channels from Sentinel-1 were compared to the NDVI derived from Sentinel-2 imagery for a 5-year period, from 2017 to 2021. The correlations between the series were studied both during and outside the growing season of the crops. Additionally, the use of the ratio of the two coherences measured at both polarimetric channels was explored. The results show that the coherence is generally well correlated with the NDVI across all seasons. The ratio between coherences at each channel is a potential alternative to the separate channels when the analysis is not restricted to the growing season of the crop, as its year-long temporal evolution more closely resembles that of the NDVI. Coherence and backscatter can be used as complementary sources of information, as backscatter-based indices describe the evolution of certain crops better than coherence.

Retrieving rice (Oryza sativa L.) net photosynthetic rate from UAV multispectral images based on machine learning methods

Photosynthesis is the key physiological activity in the process of crop growth and plays an irreplaceable role in carbon assimilation and yield formation. This study extracted rice (Oryza sativa L.) canopy reflectance based on the UAV multispectral images and analyzed the correlation between 25 vegetation indices (VIs), three textural indices (TIs), and net photosynthetic rate (Pn) at different growth stages. Linear regression (LR), support vector regression (SVR), gradient boosting decision tree (GBDT), random forest (RF), and multilayer perceptron neural network (MLP) models were employed for Pn estimation, and the modeling accuracy was compared under the input condition of VIs, VIs combined with TIs, and fusion of VIs and TIs with plant height (PH) and SPAD. The results showed that VIs and TIs generally had the relatively best correlation with Pn at the jointing-booting stage and the number of VIs with significant correlation (p< 0.05) was the largest. Therefore, the employed models could achieve the highest overall accuracy [coefficient of determination (R ²) of 0.383-0.938]. However, as the growth stage progressed, the correlation gradually weakened and resulted in accuracy decrease (R ² of 0.258-0.928 and 0.125-0.863 at the heading-flowering and ripening stages, respectively). Among the tested models, GBDT and RF models could attain the best performance based on only VIs input (with R ² ranging from 0.863 to 0.938 and from 0.815 to 0.872, respectively). Furthermore, the fusion input of VIs, TIs with PH, and SPAD could more effectively improve the model accuracy (R ² increased by 0.049-0.249, 0.063-0.470, and 0.113-0.471, respectively, for three growth stages) compared with the input combination of VIs and TIs (R ² increased by 0.015-0.090, 0.001-0.139, and 0.023-0.114). Therefore, the GBDT and RF model with fused input could be highly recommended for rice Pn estimation and the methods could also provide reference for Pn monitoring and further yield prediction at field scale.

Hydroclimate and vegetation variability of high Andean ecosystems

Mountain ecosystems are sensitive to climate fluctuations; however, the scarcity of instrumental data makes necessary the use of complementary information to study the effect of climate change on these systems. Remote sensing permits studying the dynamics of vegetation productivity and wetlands in response to climate variability at different scales. In this study we identified the main climate variables that control vegetation dynamics and water balance in Cumbres Calchaquíes, NW Argentina. For this, we built annual time series from 1986 to 2019 of Soil Adjusted Vegetation Index (SAVI, to quantify spare vegetation productivity), lake area, and snow-ice cover of peatlands, as indicators of mountain productivity and hydrology. We used a decompose function to explore trend, seasonality and random signal of the three-time series, and explored for significant changes in the mean value of consecutive periods. We used correlational analysis to explore their associations with climate records at local, regional, and global scales. The results showed that, SAVI and hydrological indicators presented different fluctuation patterns more pronounced since 2012, when they showed divergent trends with increasing SAVI and decreasing lake area and snow-ice cover. The three indicators responded differently to climate; SAVI increased in warmer years and lake area reflected the water balance of previous years. Snow-ice cover of peatlands was highly correlated with lake area. La Niña had a positive effect on lake area and snow-ice cover and a negative on SAVI, while El Niño had a negative effect on SAVI. Fluctuations of lake areas were synchronized with lake area in the nearby Argentinian puna, suggesting that climate signals have regional extent. The information provided by the three hydroclimate indicators is complementary and reflects different climate components and processes; biological processes (SAVI), physical processes (snow ice cover) and their combination (lake area). This study provides a systematic accessible replicable tool for mountain eco-hydrology long-term monitoring.

Rapid Identification of Main Vegetation Types in the Lingkong Mountain Nature Reserve Based on Multi-Temporal Modified Vegetation Indices

Nature reserves are among the most bio-diverse regions worldwide, and rapid and accurate identification is a requisite for their management. Based on the multi-temporal Sentinel-2 dataset, this study presents three multi-temporal modified vegetation indices (the multi-temporal modified normalized difference Quercus wutaishanica index (MTM-NDQI), the multi-temporal modified difference scrub grass index (MTM-DSI), and the multi-temporal modified ratio shaw index (MTM-RSI)) to improve the classification accuracy of the remote sensing of vegetation in the Lingkong Mountain Nature Reserve of China (LMNR). These three indices integrate the advantages of both the typical vegetation indices and the multi-temporal remote sensing data. By using the proposed indices with a uni-temporal modified vegetation index (the uni-temporal modified difference pine-oak mixed forest index (UTM-DMI)) and typical vegetation indices (e.g., the ratio vegetation index (RVI), the difference vegetation index (DVI), and the normalized difference vegetation index (NDVI)), an optimal feature set is obtained that includes the NDVI of December, the NDVI of April, and the UTM-DMI, MTM-NDQI, MTM-DSI, and MTM-RSI. The overall accuracy (OA) of the random forest classification (98.41%) and Kappa coefficient of the optimal feature set (0.98) were higher than those of the time series NDVI (OA = 96.03%, Kappa = 0.95), the time series RVI (OA = 95.56%, Kappa = 0.95), and the time series DVI (OA = 91.27%, Kappa = 0.90). The OAs of the rapid classification and the Kappa coefficient of the knowledge decision tree based on the optimal feature set were 95.56% and 0.95, respectively. Meanwhile, only three of the seven vegetation types were omitted or misclassified slightly. Overall, the proposed vegetation indices have advantages in identifying the vegetation types in protected areas.

Estimation of wheat tiller density using remote sensing data and machine learning methods

The tiller density is a key agronomic trait of winter wheat that is essential to field management and yield estimation. The traditional method of obtaining the wheat tiller density is based on manual counting, which is inefficient and error prone. In this study, we established machine learning models to estimate the wheat tiller density in the field using hyperspectral and multispectral remote sensing data. The results showed that the vegetation indices related to vegetation cover and leaf area index are more suitable for tiller density estimation. The optimal mean relative error for hyperspectral data was 5.46%, indicating that the results were more accurate than those for multispectral data, which had a mean relative error of 7.71%. The gradient boosted regression tree (GBRT) and random forest (RF) methods gave the best estimation accuracy when the number of samples was less than around 140 and greater than around 140, respectively. The results of this study support the extension of the tested methods to the large-scale monitoring of tiller density based on remote sensing data.

Accuracy of Vegetation Indices in Assessing Different Grades of Grassland Desertification from UAV

Grassland desertification has become one of the most serious environmental problems in the world. Grasslands are the focus of desertification research because of their ecological vulnerability. Their application on different grassland desertification grades remains limited. Therefore, in this study, 19 vegetation indices were calculated for 30 unmanned aerial vehicle (UAV) visible light images at five grades of grassland desertification in the Mu Us Sandy. Fractional Vegetation Coverage (FVC) with high accuracy was obtained through Support Vector Machine (SVM) classification, and the results were used as the reference values. Based on the FVC, the grassland desertification grades were divided into five grades: severe (FVC < 5%), high (FVC: 5-20%), moderate (FVC: 21-50%), slight (FVC: 51-70%), and non-desertification (FVC: 71-100%). The accuracy of the vegetation indices was assessed by the overall accuracy (OA), the kappa coefficient (k), and the relative error (RE). Our result showed that the accuracy of SVM-supervised classification was high in assessing each grassland desertification grade. Excess Green Red Blue Difference Index (EGRBDI), Visible Band Modified Soil Adjusted Vegetation Index (V-MSAVI), Green Leaf Index (GLI), Color Index of Vegetation Vegetative (CIVE), Red Green Blue Vegetation Index (RGBVI), and Excess Green (EXG) accurately assessed grassland desertification at severe, high, moderate, and slight grades. In addition, the Red Green Ratio Index (RGRI) and Combined 2 (COM₂) were accurate in assessing severe desertification. The assessment of the 19 indices of the non-desertification grade had low accuracy. Moreover, our result showed that the accuracy of SVM-supervised classification was high in assessing each grassland desertification grade. This study emphasizes that the applicability of the vegetation indices varies with the degree of grassland desertification and hopes to provide scientific guidance for a more accurate grassland desertification assessment.

Comparative assessment of satellite- and drone-based vegetation indices to predict arthropod biomass in shrub-steppes

Arthropod biomass is a key element in ecosystem functionality and a basic food item for many species. It must be estimated through traditional costly field sampling, normally at just a few sampling points. Arthropod biomass and plant productivity should be narrowly related because a large majority of arthropods are herbivorous, and others depend on these. Quantifying plant productivity with satellite or aerial vehicle imagery is an easy and fast procedure already tested and implemented in agriculture and field ecology. However, the capability of satellite or aerial vehicle imagery for quantifying arthropod biomass and its relationship with plant productivity has been scarcely addressed. Here, we used unmanned aerial vehicle (UAV) and satellite Sentinel-2 (S2) imagery to establish a relationship between plant productivity and arthropod biomass estimated through ground-truth field sampling in shrub steppes. We UAV-sampled seven plots of 47.6-72.3 ha at a 4-cm pixel resolution, subsequently downscaling spatial resolution to 50 cm resolution. In parallel, we used S2 imagery from the same and other dates and locations at 10-m spatial resolution. We related several vegetation indices (VIs) with arthropod biomass (epigeous, coprophagous, and four functional consumer groups: predatory, detritivore, phytophagous, and diverse) estimated at 41-48 sampling stations for UAV flying plots and in 67-79 sampling stations for S2. VIs derived from UAV were consistently and positively related to all arthropod biomass groups. Three out of seven and six out of seven S2-derived VIs were positively related to epigeous and coprophagous arthropod biomass, respectively. The blue normalized difference VI (BNDVI) and enhanced normalized difference VI (ENDVI) showed consistent and positive relationships with arthropod biomass, regardless of the arthropod group or spatial resolution. Our results showed that UAV and S2-VI imagery data may be viable and cost-efficient alternatives for quantifying arthropod biomass at large scales in shrub steppes. The relationship between VI and arthropod biomass is probably habitat-dependent, so future research should address this relationship and include several habitats to validate VIs as proxies of arthropod biomass.

Assessment of the vegetation sensitivity index in alpine meadows with a high coverage and toxic weed invasion under grazing disturbance

Maintaining healthy ecosystems is essential to ensure sustainable socio-economic development. Studies combining remote sensing data with grassland health assessments, extensively performed at different scales, are important for monitoring grassland health from a spatiotemporal perspective to enable scientific grazing management. However, most studies only use quantitative grassland degradation indices, such as grassland cover; this is done despite the fact that some degraded grasslands maintain a high level of cover solely by virtue of the proliferation of toxic weeds. Thus, seeking indices that are a more accurate representation of the health status of grassland vegetation is of utmost importance. Therefore, in order to accurately characterize the ecological integrity of grasslands (i.e., while limiting the impact of confounding variables such as weeds), we chose the grassland health comprehensive evaluation index VOR (vigor, organization, and resilience) to assess the health of grasslands on the Tibetan Plateau. We applied the VOR evaluation indices to two rangelands with different grazing intensity on the Tibetan Plateau, and extracted 11 commonly used vegetation indices based on remote sensing images of rangelands,then modeled them with the data from field surveys. Our results show that the FVC, PS, and VOR were higher in lightly grazed pastures than in heavily grazed pastures in the 2017 and 2018 growing seasons. At the beginning of the sampling period, Poaceae accounted for a greater proportion in the HG pasture. However, by August 2018, the proportion of Poaceae in the LG pasture exceeded that in the HG pasture. the proportion of Forbs in the HG pasture was significantly greater than that in the LG pasture. This indicates that vegetation response to grazing disturbance is not only a volume reduction but also a vegetation composition change. The ratio vegetation index was the most sensitive to the vegetation health response, enabling the quantification and prediction of regional vegetation health and objectively reflecting the actual condition of the grassland ecosystem. According to a multiple regression analysis, the main climatic limiting factor in the region is precipitation, which positively correlated with VOR; whereas, grazing disturbance is an important driving factor, and it is inversely correlated with VOR.

Monitoring and Identification of Agricultural Crops through Multitemporal Analysis of Optical Images and Machine Learning Algorithms

The information about where crops are distributed is useful for agri-environmental assessments, but is chiefly important for food security and agricultural policy managers. The quickness with which this information becomes available, especially over large areas, is important for decision makers. Methodologies have been proposed for the study of crops. Most of them require field survey for ground truth data and a single crop map is generated for the whole season at the end of the crop cycle and for the next crop cycle a new field survey is necessary. Here, we present models for recognizing maize (Zea mays L.), beans (Phaseolus vulgaris L.), and alfalfa (Medicago sativa L.) before the crop cycle ends without current-year field survey for ground truth data. The models were trained with an exhaustive field survey at plot level in a previous crop cycle. The field surveys begin since days before the emergence of crops to maturity. The algorithms used for classification were support vector machine (SVM) and bagged tree (BT), and the spectral information captured in the visible, red-edge, near infrared, and shortwave infrared regions bands of Sentinel 2 images was used. The models were validated within the next crop cycle each fifteen days before the mid-season. The overall accuracies range from 71.9% (38 days after the begin of cycle) to 87.5% (81 days after the begin cycle) and a kappa coefficient ranging from 0.53 at the beginning to 0.74 at mid-season.

Analysis of Biophysical Variables in an Onion Crop (Allium cepa L.) with Nitrogen Fertilization by Sentinel-2 Observations

The production of onions bulbs (Allium cepa L.) requires a high amount of nitrogen. According to the demand of sustainable agriculture, the information-development and communication technologies allow for improving the efficiency of nitrogen fertilization. In the south of the province of Buenos Aires, Argentina, between 8000 and 10,000 hectares per year^-1 are cultivated in the districts of Villarino and Patagones. This work aimed to analyze the relationship of biophysical variables: leaf area index (LAI), canopy chlorophyll content (CCC), and canopy cover factor (fCOVER), with the nitrogen fertilization of an intermediate cycle onion crop and its effects on yield. A field trial study with different doses of granulated urea and granulated urea was carried out, where biophysical characteristics were evaluated in the field and in Sentinel-2 satellite observations. Field data correlated well with satellite data, with an R² of 0.91, 0.96, and 0.85 for LAI, fCOVER, and CCC, respectively. The application of nitrogen in all its doses produced significantly higher yields than the control. The LAI and CCC variables had a positive correlation with yield in the months of November and December. A significant difference was observed between U250 (62 Mg ha^-1) and the other treatments. The U500 dose led to a yield increase of 27% compared to U250, while the difference between U750 and U500 was 6%.

Research on Cyanobacterial-Bloom Detection Based on Multispectral Imaging and Deep-Learning Method

Frequent outbreaks of cyanobacterial blooms have become one of the most challenging water ecosystem issues and a critical concern in environmental protection. To overcome the poor stability of traditional detection algorithms, this paper proposes a method for detecting cyanobacterial blooms based on a deep-learning algorithm. An improved vegetation-index method based on a multispectral image taken by an Unmanned Aerial Vehicle (UAV) was adopted to extract inconspicuous spectral features of cyanobacterial blooms. To enhance the recognition accuracy of cyanobacterial blooms in complex scenes with noise such as reflections and shadows, an improved transformer model based on a feature-enhancement module and pixel-correction fusion was employed. The algorithm proposed in this paper was implemented in several rivers in China, achieving a detection accuracy of cyanobacterial blooms of more than 85%. The estimate of the proportion of the algae bloom contamination area and the severity of pollution were basically accurate. This paper can lay a foundation for ecological and environmental departments for the effective prevention and control of cyanobacterial blooms.

Prediction of Chlorophyll Content in Multi-Temporal Winter Wheat Based on Multispectral and Machine Learning

To obtain the canopy chlorophyll content of winter wheat in a rapid and non-destructive high-throughput manner, the study was conducted on winter wheat in Xinjiang Manas Experimental Base in 2021, and the multispectral images of two water treatments' normal irrigation (NI) and drought stress (DS) in three key fertility stages (heading, flowering, and filling) of winter wheat were obtained by DJI P4M unmanned aerial vehicle (UAV). The flag leaf chlorophyll content (CC) data of different genotypes in the field were obtained by SPAD-502 Plus chlorophyll meter. Firstly, the CC distribution of different genotypes was studied, then, 13 vegetation indices, combined with the Random Forest algorithm and correlation evaluation of CC, and 14 vegetation indices were used for vegetation index preference. Finally, preferential vegetation indices and nine machine learning algorithms, Ridge regression with cross-validation (RidgeCV), Ridge, Adaboost Regression, Bagging_Regressor, K_Neighbor, Gradient_Boosting_Regressor, Random Forest, Support Vector Machine (SVM), and Least absolute shrinkage and selection operator (Lasso), were preferentially selected to construct the CC estimation models under two water treatments at three different fertility stages, which were evaluated by correlation coefficient (r), root means square error (RMSE) and the normalized root mean square error (NRMSE) to select the optimal estimation model. The results showed that the CC values under normal irrigation were higher than those underwater limitation treatment at different fertility stages; several vegetation indices and CC values showed a highly significant correlation, with the highest correlation reaching.51; in the prediction model construction of CC values, different models under normal irrigation and water limitation treatment had high estimation accuracy, among which the model with the highest prediction accuracy under normal irrigation was at the heading stage. The highest precision of the model prediction under normal irrigation was in the RidgeCV model (r = 0.63, RMSE = 3.28, NRMSE = 16.2%) and the highest precision of the model prediction under water limitation treatment was in the SVM model (r = 0.63, RMSE = 3.47, NRMSE = 19.2%).

Non-invasive Estimation of Foliar Nitrogen Concentration Using Spectral Characteristics of Menthol Mint (Mentha arvensis L.)

Menthol mint (Mentha arvensis L., Family: Lamiaceae), popularly known as corn mint or Japanese mint, is an important industrial crop that is widely grown for its valued essential oil. Nitrogen (N) is an important macro-nutrient and an essential factor for optimizing the yield and quality of crops. Hence, rapid and accurate estimation of the N content is crucial for nutrient diagnosis in plants and to make precise N fertilizer recommendations. Generally, N concentration is estimated by destructive sampling methods; however, an indirect assessment may be possible based on spectral characteristics. This study aimed to compare the foliar N concentration based on non-destructive (reflectance) and destructive (laboratory analyses) methods in menthol mint. Foliar N concentration was measured through the Kjeldahl method and reflectance by Miniature Leaf Spectrometer C-710 (CID Bio-Science). Using reflectance data, several vegetation indices (VIs), that is, normalized difference red edge (NDRE), red edge normalized difference vegetation index (reNDVI), simple ratio (SR), green-red vegetation index (GRVI), canopy chlorophyll content index (CCCI), photochemical reflectance index (PRI), green chlorophyll index (CI Green ), red edge chlorophyll index (CI Red Edge ), canopy chlorophyll index (CCI), normalized pigment chlorophyll ratio index (NPCI), and structure insensitive pigment index (SIPI), were developed to determine the foliar N concentration. The highest correlation (r) between VIs and foliar N concentrations was achieved by NDRE (0.89), followed by reNDVI (0.84), SR (0.83), GRVI (0.78), and CCCI (0.76). Among the VIs, the NDRE index has been found to be the most accurate index that can precisely predict the foliar N concentration (R 2 = 0.79, RMSE = 0.18). In summary, the N deficiencies faced by the crop during its growth period can be detected effectively by calculating NDRE and reNDVI, which can be used as indicators for recommending precise management strategies for the application of nitrogenous fertilizers.

Characteristics of Picea neoveitchii tree growth in mountain areas of central China: insights from isotopic compositions and satellite-derived indices

Leaf nitrogen (N) status and stable isotope ratios of carbon (δ¹³C) and nitrogen (δ¹⁵N) were used to study environmental factors that control mountain individuals of Picea neoveitchii trees, a coniferous species endemic and endangered in China. From May to September 2016, we carried out observations at four different altitude locations extending southeast of Daba Mountain in western Hubei Province. Needle-shaped leaf δ¹³C was positively correlated with needle N and C content calculated from the needle area (N_area and C_area content), needle δ¹⁵N, needle mass, and leaf mass per area (LMA), respectively. Needle δ¹⁵N was also positively correlated with monthly temperature and precipitation for the current month and last month. The seasonal normalised difference vegetation index (NDVI) was highest in June at the lowest altitude and August at the highest altitude. We found that N availability as an important driving factor for tree growth is controlled by surface soil temperature, while in summer, air temperatures above 23 °C exceed the physiological threshold of trees and limit the growth of trees. We concluded that the negative effect of higher temperature on tree growth is greater than the positive effect of higher nitrogen.

Applications of a Hyperspectral Imaging System Used to Estimate Wheat Grain Protein: A Review

Recent research advances in wheat have focused not only on increasing grain yields, but also on establishing higher grain quality. Wheat quality is primarily determined by the grain protein content (GPC) and composition, and both of these are affected by nitrogen (N) levels in the plant as it develops during the growing season. Hyperspectral remote sensing is gradually becoming recognized as an economical alternative to traditional destructive field sampling methods and laboratory testing as a means of determining the N status within wheat. Currently, hyperspectral vegetation indices (VIs) and linear nonparametric regression are the primary tools for monitoring the N status of wheat. Machine learning algorithms have been increasingly applied to model the nonlinear relationship between spectral data and wheat N status. This study is a comprehensive review of available N-related hyperspectral VIs and aims to inform the selection of VIs under field conditions. The combination of feature mining and machine learning algorithms is discussed as an application of hyperspectral imaging systems. We discuss the major challenges and future directions for evaluating and assessing wheat N status. Finally, we suggest that the underlying mechanism of protein formation in wheat grains as determined by using hyperspectral imaging systems needs to be further investigated. This overview provides theoretical and technical support to promote applications of hyperspectral imaging systems in wheat N status assessments; in addition, it can be applied to help monitor and evaluate food and nutrition security.

Evaluating the accuracy of ARMA and multi-index methods for predicting winter wheat maturity date

BACKGROUND

Accurate and timely prediction of regional winter wheat maturity date can provide essential information to improve the management of agriculture and avoid declines in the yield and quality of crops. In this paper, we propose the use of an autoregressive moving-average model to predict vegetation indices on 1, 9, and 17 May each year, and applied them to the methods of evaluating crop maturity based on vegetation indices. Growing degree days and a widely applied local empirical method were selected to explore and compare the feasibility of several methods. We analyzed winter wheat harvested from the Guanzhong Plain during 2003-2013 and used leave-one-out cross-validation to compare and verify the performance of the maturity prediction methods.

RESULTS

The results demonstrated that (i) the vegetation index methods and growing degree days methods predicted maturity with higher accuracy than did the widely applied local empirical method, and (ii) the two-step filtering method based on future meteorological data from The Observing System Research and Predictability Experiment Interactive Grand Global Ensemble exhibited the highest prediction accuracy on 1 May and had the lowest error fluctuation range on 17 May.

CONCLUSION

These results provide new insights for predicting regional crop maturity, deploying agricultural harvesting equipment in various regions, and avoiding decreases in crop yields caused by adverse weather. © 2021 Society of Chemical Industry.

Reduced ecosystem resilience quantifies fine-scale heterogeneity in tropical forest mortality responses to drought

Sensitivity of forest mortality to drought in carbon-dense tropical forests remains fraught with uncertainty, while extreme droughts are predicted to be more frequent and intense. Here, the potential of temporal autocorrelation of high-frequency variability in Landsat Enhanced Vegetation Index (EVI), an indicator of ecosystem resilience, to predict spatial and temporal variations of forest biomass mortality is evaluated against in situ census observations for 64 site-year combinations in Costa Rican tropical dry forests during the 2015 ENSO drought. Temporal autocorrelation, within the optimal moving window of 24 months, demonstrated robust predictive power for in situ mortality (leave-one-out cross-validation R²  = 0.54), which allows for estimates of annual biomass mortality patterns at 30 m resolution. Subsequent spatial analysis showed substantial fine-scale heterogeneity of forest mortality patterns, largely driven by drought intensity and ecosystem properties related to plant water use such as forest deciduousness and topography. Highly deciduous forest patches demonstrated much lower mortality sensitivity to drought stress than less deciduous forest patches after elevation was controlled. Our results highlight the potential of high-resolution remote sensing to "fingerprint" forest mortality and the significant role of ecosystem heterogeneity in forest biomass resistance to drought.

Higher plant photosynthetic capability in autumn responding to low atmospheric vapor pressure deficit

It has been long established that the terrestrial vegetation in spring has stronger photosynthetic capability than in autumn. However, this study challenges this consensus by comparing photosynthetic capability of terrestrial vegetation between the spring and autumn seasons based on measurements of 100 in situ eddy covariance towers over global extratropical ecosystems. At the majority of these sites, photosynthetic capability, indicated by light use efficiency (LUE) and apparent quantum efficiency, is significantly higher in autumn than in spring, due to lower atmosphere vapor pressure deficit (VPD) at the same air temperature. Seasonal VPD differences also substantially explain the interannual variability of the differences in photosynthetic capability between spring and autumn. We further reveal that VPD in autumn is significantly lower than in spring over 74.14% of extratropical areas, based on a global climate dataset. In contrast, LUE derived from a data-driven vegetation production dataset is significantly higher in autumn in over 61.02% of extratropical vegetated areas. Six Earth system models consistently projected continuous larger VPD values in spring compared with autumn, which implies that the impacts on vegetation growth will long exist and should be adequately considered when assessing the seasonal responses of terrestrial ecosystems to future climate conditions.

•

Autumn VPD is lower than spring VPD at the same air temperature over majority of the extratropical vegetated land

•

Photosynthetic capability is significantly higher in autumn than in spring due to lower VPD

•

Earth System Models projected continuous larger VPD values in spring as against autumn

Potential Use of Hyperspectral Reflectance as a High-Throughput Nondestructive Phenotyping Tool for Assessing Salt Tolerance in Advanced Spring Wheat Lines under Field Conditions

The incorporation of stress tolerance indices (STIs) with the early estimation of grain yield (GY) in an expeditious and nondestructive manner can enable breeders for ensuring the success of genotype development for a wide range of environmental conditions. In this study, the relative performance of GY for sixty-four spring wheat germplasm under the control and 15.0 dS m^-1 NaCl were compared through different STIs, and the ability of a hyperspectral reflectance tool for the early estimation of GY and STIs was assessed using twenty spectral reflectance indices (SRIs; 10 vegetation SRIs and 10 water SRIs). The results showed that salinity treatments, genotypes, and their interactions had significant effects on the GY and nearly all SRIs. Significant genotypic variations were also observed for all STIs. Based on the GY under the control (GYc) and salinity (GYs) conditions and all STIs, the tested genotypes were classified into three salinity tolerance groups (salt-tolerant, salt-sensitive, and moderately salt-tolerant groups). Most vegetation and water SRIs showed strong relationships with the GYc, stress tolerance index (STI), and geometric mean productivity (GMP); moderate relationships with GYs and sometimes with the tolerance index (TOL); and weak relationships with the yield stability index (YSI) and stress susceptibility index (SSI). Obvious differences in the spectral reflectance curves were found among the three salinity tolerance groups under the control and salinity conditions. Stepwise multiple linear regressions identified three SRIs from each vegetation and water SRI as the most influential indices that contributed the most variation in the GY. These SRIs were much more effective in estimating the GYc (R² = 0.64 - 0.79) than GYs (R² = 0.38 - 0.47). They also provided a much accurate estimation of the GYc and GYs for the moderately salt-tolerant genotype group; YSI, SSI, and TOL for the salt-sensitive genotypes group; and STI and GMP for all the three salinity tolerance groups. Overall, the results of this study highlight the potential of using a hyperspectral reflectance tool in breeding programs for phenotyping a sufficient number of genotypes under a wide range of environmental conditions in a cost-effective, noninvasive, and expeditious manner. This will aid in accelerating the development of genotypes for salinity conditions in breeding programs.

Hyperspectral Characteristics of an Individual Leaf of Wheat Grown under Nitrogen Gradient

Since the application of hyperspectral technology to agriculture, many scientists have been conducting studies to apply the technology in crop diagnosis. However, due to the properties of optical devices, the reflectances obtained according to the image acquisition conditions are different. Nevertheless, there is no optimized method for minimizing such technical errors in applying hyperspectral imaging. Therefore, this study was conducted to find the appropriate image acquisition conditions that reflect the growth status of wheat grown under different nitrogen fertilization regimes. The experiment plots were comprised of six plots with various N application levels of 145.6 kg N ha^-1 (N1), 109.2 kg N ha^-1 (N2), 91.0 kg N ha^-1 (N3), 72.8 kg N ha^-1 (N4), 54.6 kg N ha^-1 (N5), and 36.4 kg N ha^-1 (N6). Hyperspectral image acquisitions were performed at different shooting angles of 105° and 125° from the surface, and spike, flag leaf, and the second uppermost leaf were divided into five parts from apex to base when analyzing the images. The growth analysis conducted at heading showed that the N6 was 85.6% in the plant height, 44.1% in LAI, and 64.9% in SPAD as compared to N1. The nitrogen content in the leaf decreased by 55.2% compared to N1 and the quantity was 44.9% in N6 compared to N1. Based on the vegetation indices obtained from hyperspectral reflectances at the heading stage, the spike was not suitable for analysis. In the case of the flag leaf and the 2nd uppermost leaf, the vegetation indices from spectral data taken at 105 degrees were more appropriate for acquiring imaging data by clearly dividing the effects of fertilization level. The results of the regional variation in a leaf showed that the region of interest (ROI), which is close to the apex of the flag leaf and the base of the second uppermost leaf, has a high coefficient of determination between the fertilization levels and the vegetation indices, which effectively reflected the status of wheat.

Analysis of New RGB Vegetation Indices for PHYVV and TMV Identification in Jalapeño Pepper (Capsicum annuum) Leaves Using CNNs-Based Model

Recently, deep-learning techniques have become the foundations for many breakthroughs in the automated identification of plant diseases. In the agricultural sector, many recent visual-computer approaches use deep-learning models. In this approach, a novel predictive analytics methodology to identify Tobacco Mosaic Virus (TMV) and Pepper Huasteco Yellow Vein Virus (PHYVV) visual symptoms on Jalapeño pepper (Capsicum annuum L.) leaves by using image-processing and deep-learning classification models is presented. The proposed image-processing approach is based on the utilization of Normalized Red-Blue Vegetation Index (NRBVI) and Normalized Green-Blue Vegetation Index (NGBVI) as new RGB-based vegetation indices, and its subsequent Jet pallet colored version NRBVI-Jet NGBVI-Jet as pre-processing algorithms. Furthermore, four standard pre-trained deep-learning architectures, Visual Geometry Group-16 (VGG-16), Xception, Inception v3, and MobileNet v2, were implemented for classification purposes. The objective of this methodology was to find the most accurate combination of vegetation index pre-processing algorithms and pre-trained deep- learning classification models. Transfer learning was applied to fine tune the pre-trained deep- learning models and data augmentation was also applied to prevent the models from overfitting. The performance of the models was evaluated using Top-1 accuracy, precision, recall, and F1-score using test data. The results showed that the best model was an Xception-based model that uses the NGBVI dataset. This model reached an average Top-1 test accuracy of 98.3%. A complete analysis of the different vegetation index representations using models based on deep-learning architectures is presented along with the study of the learning curves of these deep-learning models during the training phase.

Inversion of Winter Wheat Growth Parameters and Yield Under Different Water Treatments Based on UAV Multispectral Remote Sensing

In recent years, the unmanned aerial vehicle (UAV) remote sensing system has been rapidly developed and applied in accurate estimation of crop parameters and yield at farm scale. To develop the major contribution of UAV multispectral images in predicting winter wheat leaf area index (LAI), chlorophyll content (called soil and plant analyzer development [SPAD]), and yield under different water treatments (low water level, medium water level, and high water level), vegetation indices (VIs) originating from UAV multispectral images were used during key winter wheat growth stages. The estimation performances of the models (linear regression, quadratic polynomial regression, and exponential and multiple linear regression models) on the basis of VIs were compared to get the optimal prediction method of crop parameters and yield. Results showed that LAI and SPAD derived from VIs both had high correlations compared with measured data, with determination coefficients of 0.911 and 0.812 (multivariable regression [MLR] model, normalized difference VI [NDVI], soil adjusted VI [SAVI], enhanced VI [EVI], and difference VI [DVI]), 0.899 and 0.87 (quadratic polynomial regression, NDVI), and 0.749 and 0.829 (quadratic polynomial regression, NDVI) under low, medium, and high water levels, respectively. The LAI and SPAD derived from VIs had better potential in estimating winter wheat yield by using multivariable linear regressions, compared to the estimation yield based on VIs directly derived from UAV multispectral images alone by using linear regression, quadratic polynomial regression, and exponential models. When crop parameters (LAI and SPAD) in the flowering period were adopted to estimate yield by using multiple linear regressions, a high correlation of 0.807 was found, while the accuracy was over 87%. Importing LAI and SPAD obtained from UAV multispectral imagery based on VIs into the yield estimation model could significantly enhance the estimation performance. This study indicates that the multivariable linear regression could accurately estimate winter wheat LAI, SPAD, and yield under different water treatments, which has a certain reference value for the popularization and application of UAV remote sensing in precision agriculture.

Molecular Mapping of Water-Stress Responsive Genomic Loci in Lettuce (Lactuca spp.) Using Kinetics Chlorophyll Fluorescence, Hyperspectral Imaging and Machine Learning

Deep understanding of genetic architecture of water-stress tolerance is critical for efficient and optimal development of water-stress tolerant cultivars, which is the most economical and environmentally sound approach to maintain lettuce production with limited irrigation. Lettuce (Lactuca sativa L.) production in areas with limited precipitation relies heavily on the use of ground water for irrigation. Lettuce plants are highly susceptible to water-stress, which also affects their nutrient uptake efficiency. Water stressed plants show reduced growth, lower biomass, and early bolting and flowering resulting in bitter flavors. Traditional phenotyping methods to evaluate water-stress are labor intensive, time-consuming and prone to errors. High throughput phenotyping platforms using kinetic chlorophyll fluorescence and hyperspectral imaging can effectively attain physiological traits related to photosynthesis and secondary metabolites that can enhance breeding efficiency for water-stress tolerance. Kinetic chlorophyll fluorescence and hyperspectral imaging along with traditional horticultural traits identified genomic loci affected by water-stress. Supervised machine learning models were evaluated for their accuracy to distinguish water-stressed plants and to identify the most important water-stress related parameters in lettuce. Random Forest (RF) had classification accuracy of 89.7% using kinetic chlorophyll fluorescence parameters and Neural Network (NN) had classification accuracy of 89.8% using hyperspectral imaging derived vegetation indices. The top ten chlorophyll fluorescence parameters and vegetation indices selected by sequential forward selection by RF and NN were genetically mapped using a L. sativa × L. serriola interspecific recombinant inbred line (RIL) population. A total of 25 quantitative trait loci (QTL) segregating for water-stress related horticultural traits, 26 QTL for the chlorophyll fluorescence traits and 34 QTL for spectral vegetation indices (VI) were identified. The percent phenotypic variation (PV) explained by the horticultural QTL ranged from 6.41 to 19.5%, PV explained by chlorophyll fluorescence QTL ranged from 6.93 to 13.26% while the PV explained by the VI QTL ranged from 7.2 to 17.19%. Eight QTL clusters harboring co-localized QTL for horticultural traits, chlorophyll fluorescence parameters and VI were identified on six lettuce chromosomes. Molecular markers linked to the mapped QTL clusters can be targeted for marker-assisted selection to develop water-stress tolerant lettuce.

Advances in Unmanned Aerial System Remote Sensing for Precision Viticulture

New technologies for management, monitoring, and control of spatio-temporal crop variability in precision viticulture scenarios are numerous. Remote sensing relies on sensors able to provide useful data for the improvement of management efficiency and the optimization of inputs. unmanned aerial systems (UASs) are the newest and most versatile tools, characterized by high precision and accuracy, flexibility, and low operating costs. The work aims at providing a complete overview of the application of UASs in precision viticulture, focusing on the different application purposes, the applied equipment, the potential of technologies combined with UASs for identifying vineyards' variability. The review discusses the potential of UASs in viticulture by distinguishing five areas of application: rows segmentation and crop features detection techniques; vineyard variability monitoring; estimation of row area and volume; disease detection; vigor and prescription maps creation. Technological innovation and low purchase costs make UASs the core tools for decision support in the customary use by winegrowers. The ability of the systems to respond to the current demands for the acquisition of digital technologies in agricultural fields makes UASs a candidate to play an increasingly important role in future scenarios of viticulture application.

Estimation of Peanut Leaf Area Index from Unmanned Aerial Vehicle Multispectral Images

Leaf area index (LAI) is used to predict crop yield, and unmanned aerial vehicles (UAVs) provide new ways to monitor LAI. In this study, we used a fixed-wing UAV with multispectral cameras for remote sensing monitoring. We conducted field experiments with two peanut varieties at different planting densities to estimate LAI from multispectral images and establish a high-precision LAI prediction model. We used eight vegetation indices (VIs) and developed simple regression and artificial neural network (BPN) models for LAI and spectral VIs. The empirical model was calibrated to estimate peanut LAI, and the best model was selected from the coefficient of determination and root mean square error. The red (660 nm) and near-infrared (790 nm) bands effectively predicted peanut LAI, and LAI increased with planting density. The predictive accuracy of the multiple regression model was higher than that of the single linear regression models, and the correlations between Modified Red-Edge Simple Ratio Index (MSR), Ratio Vegetation Index (RVI), Normalized Difference Vegetation Index (NDVI), and LAI were higher than the other indices. The combined VI BPN model was more accurate than the single VI BPN model, and the BPN model accuracy was higher. Planting density affects peanut LAI, and reflectance-based vegetation indices can help predict LAI.

Evaluation of Vegetation Index-Based Curve Fitting Models for Accurate Classification of Salt Marsh Vegetation Using Sentinel-2 Time-Series

The successful launch of the Sentinel-2 constellation satellite, along with advanced cloud detection algorithms, has enabled the generation of continuous time series at high spatial and temporal resolutions, which is in turn expected to enable the classification of salt marsh vegetation over larger spatiotemporal scales. This study presents a critical comparison of vegetation index (VI) and curve fitting methods-two key factors for time series construction that potentially influence vegetation classification performance. To accomplish this objective, the stability of five different VI time series, namely Normalized Difference Vegetation Index (NDVI), Soil-Adjusted Vegetation Index (SAVI), Enhanced Vegetation Index (EVI), Green Normalized Difference Vegetation Index (GNDVI), and Water-Adjusted Vegetation Index (WAVI), was compared empirically; the suitability between three curve fitting methods, namely Asymmetric Gaussian (AG), Double Logistic (DL), and Two-term Fourier (TF), and VI time series was measured using the coefficient of determination, and the salt marsh vegetation separability among different combinations of VI time series and curve fitting methods (i.e., VI time series-based curve fitting model) was quantified using overall the Jeffries-Matusita distance. Six common types of salt marsh vegetation from three typical coastal sites in China were used to validate these findings, which demonstrate: (1) the SAVI performed best in terms of time series stability, while the EVI exhibited relatively poor time series stability with conspicuous outliers induced by the sensitivity to omitted clouds and shadows; (2) the DL method commonly resulted in the most accurate classification of different salt marsh vegetation types, especially when combined with the EVI time series, followed by the TF method; and (3) the SAVI/NDVI-based DL/TF model demonstrated comparable efficiency for classifying salt marsh vegetation. Notably, the SAVI/NDVI-based DL model performed most strongly for high latitude regions with a continental climate, whilst the SAVI/NDVI-based TF model appears to be better suited to mid- to low latitude regions dominated by a monsoonal climate.

Data-Driven Methods for the Estimation of Leaf Water and Dry Matter Content: Performances, Potential and Limitations

Leaf equivalent water thickness (EWT) and dry matter content (expressed as leaf mass per area (LMA)) are two critical traits for vegetation function monitoring, crop yield estimation, and precise agriculture management. Data-driven methods are widely used for remote sensing of leaf EWT and LMA because of their simplicity, satisfactory accuracy, and computation efficiency, such as the vegetation indices (VI)-based and machine learning (ML)-based methods. However, most of the data-driven methods are utilized at the canopy level, comparison of the performances of the data-driven methods at the leaf level has not been well documented. Moreover, the ML-based data-driven methods generally adopt leaf optical properties directly as their inputs, which may subsequently decrease their ability in remote sensing of leaf biochemical constituents. Performances of the ML-based methods cooperating with VI are rarely evaluated. Using the independent LOPEX and ANGERS datasets, we compared the performances of three data-driven methods: VI-based, ML-reflectance-based, and ML-VI-based methods, for the estimation of leaf EWT and LMA. Three sampling strategies were also utilized for evaluation of the generalization of these data-driven methods. Our results evidenced that ML-VI-based methods were the most accurate among these data-driven methods. Compared to the ML-reflectance-based and VI-based methods, the ML-VI-based model with support vector regression overall reduced errors by 5.7% (41.5%) and 1.8% (12.4%) for the estimation of leaf EWT (LMA), respectively. The ML-VI-based model inherits advantages of vegetation indices and ML techniques, which made it sensitive to changes of leaf biochemical constituents and capable of solving nonlinear tasks. It is thus recommended for the estimation of EWT and LMA at the leaf level. Moreover, its performance can further be enhanced by improving its generalization ability, such as adopting techniques on the selection of better wavelengths and definition of new vegetation indices. These results thus provided a prior knowledge of the data-driven methods and can be helpful for future studies on the remote sensing of leaf biochemical constituents.

Hyperspectral Reflectance of Light-Adapted Leaves Can Predict Both Dark- and Light-Adapted Chl Fluorescence Parameters, and the Effects of Chronic Ozone Exposure on Date Palm (Phoenix dactylifera)

High-throughput and large-scale measurements of chlorophyll a fluorescence (ChlF) are of great interest to investigate the photosynthetic performance of plants in the field. Here, we tested the capability to rapidly, precisely, and simultaneously estimate the number of pulse-amplitude-modulation ChlF parameters commonly calculated from both dark- and light-adapted leaves (an operation which usually takes tens of minutes) from the reflectance of hyperspectral data collected on light-adapted leaves of date palm seedlings chronically exposed in a FACE facility to three ozone (O3) concentrations (ambient air, AA; target 1.5 × AA O3, named as moderate O3, MO; target 2 × AA O3, named as elevated O3, EO) for 75 consecutive days. Leaf spectral measurements were paired with reference measurements of ChlF, and predictive spectral models were constructed using partial least squares regression. Most of the ChlF parameters were well predicted by spectroscopic models (average model goodness-of-fit for validation, R2: 0.53-0.82). Furthermore, comparing the full-range spectral profiles (i.e., 400-2400 nm), it was possible to distinguish with high accuracy (81% of success) plants exposed to the different O3 concentrations, especially those exposed to EO from those exposed to MO and AA. This was possible even in the absence of visible foliar injury and using a moderately O3-susceptible species like the date palm. The latter view is confirmed by the few variations of the ChlF parameters, that occurred only under EO. The results of the current study could be applied in several scientific fields, such as precision agriculture and plant phenotyping.

Quantifying large-scale ecosystem stability with remote sensing data

To fully understand ecosystem functioning under global change, we need to be able to measure the stability of ecosystem functioning at multiple spatial scales. Although a number of stability components have been established at small spatial scales, there has been little progress in scaling these measures up to the landscape. Remote sensing data holds huge potential for studying processes at landscape scales but requires quantitative measures that are comparable from experimental field data to satellite remote sensing. Here we present a methodology to extract four components of ecosystem functioning stability from satellite-derived time series of Enhanced Vegetation Index (EVI) data. The four stability components are as follows: variability, resistance, recovery time and recovery rate in ecosystem functioning. We apply our method to the island of Ireland to demonstrate the use of remotely sensed data to identify large disturbance events in productivity. Our method uses stability measures that have been established at the field-plot scale to quantify the stability of ecosystem functioning. This makes our method consistent with previous small-scale stability research, whilst dealing with the unique challenges of using remotely sensed data including noise. We encourage the use of remotely-sensed data in assessing the stability of ecosystems at a scale that is relevant to conservation and management practices.

Simulation and Verification of Vertical Heterogeneity Spectral Response of Winter Wheat Based on the mSCOPE Model

Vertical heterogeneity of the biochemical characteristics of crop canopy is important in diagnosing and monitoring nutrition, disease, and crop yield via remote sensing. However, the research on vertical isomerism was not comprehensive. Experiments were carried out from the two levels of simulation and verification to analyze the applicability of this recently development model. Effects of winter wheat on spectrum were studied when input different structure parameters (e.g., leaf area index (LAI)) and physicochemical parameters (e.g., chlorophyll content (Chla+b) and water content (Cw)) to the mSCOPE (Soil Canopy Observation, Photochemistry, and Energy fluxes) model. The maximum operating efficiency was 127.43, when the winter wheat was stratified into three layers. Meanwhile, the simulation results also proved that: the vertical profile of LAI had an influence on canopy reflectance in almost all bands; the vertical profile of Chla+b mainly affected the reflectivity of visible region; the vertical profile of Cw only affected the near-infrared reflectance. The verification results showed that the vegetation indexes (VIs) selected of different bands were strongly correlated with the parameters of the canopy. LAI, Chla+b and Cw affected VIs estimation related to LAI, Chla+b and Cw respectively. The Root Mean Square Error (RMSE) of the new-proposed NDVIgreen was the smallest, which was 0.05. Sensitivity analysis showed that the spectrum was more sensitive to changes in upper layer parameters, which verified the rationality of mSCOPE model in explaining the law that light penetration in vertical nonuniform canopy gradually decreases with the increase of layers.

[Model construction and application for nitrogen nutrition monitoring and diagnosis in double-cropping rice of Jiangxi Province, China]

The spectrometer-based nitrogen (N) nutrition monitoring and diagnosis models for double-cropping rice in Jiangxi is important for recommending precise N topdressing rate, achieving high yield, improving grain quality and increasing economic efficiency. Field experiments were conducted in Jiangxi in 2016 and 2017, involving different early rice and late rice cultivars and N application rates. Plant N accumulation (PNA) and canopy spectral vegetation indices (VIs) were measured at tillering and jointing stages with two spectrometers, i.e., GreenSeeker (an active multispectral sensor containing 780 and 660 nm wavelengths) and crop growth monitoring and diagnosis apparatus (CGMD, a passive multispectral sensor containing 810 and 720 nm wavelengths). The VI-based models of PNA were established from a experimental dataset and then validated using an independent dataset. The N topdressing rates for tillering and jointing stages were calculated using the newly developed N spectral diagnosis model and higher yield cultivation experience of double-cropping rice. The results showed that the VIs from two spectrometers were strongly positively correlated with PNA at both growth stages, with the model performance for tillering or jointing stages was better than that for the early growth stages. The exponential equation of normalized difference vegetation index (NDVI_(780,660)) from GreenSeeker could be used to estimate PNA with a determination coefficient (R²) in the range of 0.92-0.94, the root mean square error (RMSE), relative root mean square error (RRMSE) and correlation coefficient (r) of model validation in the range of 3.09-5.96 kg·hm^-2, 5.8%-18.5% and 0.92-0.98, respectively. The linear equation of difference vegetation index (DVI_(810,720)) from CGMD could be used to estimate PNA with a R² in the range of 0.90-0.93, the RMSE, RRMSE and r of model validation in the range of 3.71-6.33 kg·hm^-2, 11.7%-14.3% and 0.93-0.96, respectively. The recommended N topdressing rate with CGMD was higher than that with GreenSeeker. Compared with conventional farmer's plan, the precision N application plan reduced N fertilizer application rate by 5.5 kg·hm^-2, while N agronomic efficiency and net income was improved by 0.8% and 128 yuan·hm^-2, respectively. Application of the spectral monitoring and diagnosis method to guiding fertilization could reduce cost and increase grain yield and net income, and thus had great potential for guiding double-cropping rice production.

Monitoring Wheat Growth Using a Portable Three-Band Instrument for Crop Growth Monitoring and Diagnosis

An instrument developed to monitor and diagnose crop growth can quickly and non-destructively obtain crop growth information, which is helpful for crop field production and management. Focusing on the problems with existing two-band instruments used for crop growth monitoring and diagnosis, such as insufficient information available on crop growth and low accuracy of some growth indices retrieval, our research team developed a portable three-band instrument for crop-growth monitoring and diagnosis (CGMD) that obtains a larger amount of information. Based on CGMD, this paper carried out studies on monitoring wheat growth indices. According to the acquired three-band reflectance spectra, the combined indices were constructed by combining different bands, two-band vegetation indices (NDVI, RVI, and DVI), and three-band vegetation indices (TVI-1 and TVI-2). The fitting results of the vegetation indices obtained by CGMD and the commercial instrument FieldSpec HandHeld2 was high and the new instrument could be used for monitoring the canopy vegetation indices. By fitting each vegetation index to the growth index, the results showed that the optimal vegetation indices corresponding to leaf area index (LAI), leaf dry weight (LDW), leaf nitrogen content (LNC), and leaf nitrogen accumulation (LNA) were TVI-2, TVI-1, NDVI (R₇₃₀, R₈₁₅), and NDVI (R₇₃₀, R₈₁₅), respectively. R² values corresponding to LAI, LDW, LNC and LNA were 0.64, 0.84, 0.60, and 0.82, respectively, and their relative root mean square error (RRMSE) values were 0.29, 0.26, 0.17, and 0.30, respectively. The addition of the red spectral band to CGMD effectively improved the monitoring results of wheat LAI and LDW. Focusing the problem of vegetation index saturation, this paper proposed a method to construct the wheat-growth-index spectral monitoring models that were defined according to the growth periods. It improved the prediction accuracy of LAI, LDW, and LNA, with R² values of 0.79, 0.85, and 0.85, respectively, and the RRMSE values of these growth indices were 0.22, 0.23, and 0.28, respectively. The method proposed here could be used for the guidance of wheat field cultivation.

Competing Vegetation Structure Indices for Estimating Spatial Constrains in Carabid Abundance Patterns in Chinese Grasslands Reveal Complex Scale and Habitat Patterns

Carabid communities are influenced by landscape features. Chinese steppes are subject to increasing desertification processes that are changing land-cover characteristics with negative impacts on insect communities. Despite those warnings, how land-cover characteristics influence carabid communities in steppe ecosystems remains unknown. The aim of this study is to investigate how landscape characteristics drive carabid abundance in different steppes (desert, typical, and meadow steppes) at different spatial scales. Carabid abundances were estimated using pitfall traps. Various landscape indices were derived from Landsat 8 Operational Land Imager (OLI) images. Indices expressing moisture and productivity were, in general, those with the highest correlations. Different indices capture landscape aspects that influence carabid abundance at different scales, in which the patchiness of desert vegetation plays a major role. Carabid abundance correlations with landscape characteristics rely on the type of grassland, on the vegetation index, and on the scale considered. Proper scales and indices are steppe type-specific, highlighting the need of considering various scales and indices to explain species abundances from remotely sensed data.

Estimation of Crop Growth Parameters Using UAV-Based Hyperspectral Remote Sensing Data

Above-ground biomass (AGB) and the leaf area index (LAI) are important indicators for the assessment of crop growth, and are therefore important for agricultural management. Although improvements have been made in the monitoring of crop growth parameters using ground- and satellite-based sensors, the application of these technologies is limited by imaging difficulties, complex data processing, and low spatial resolution. Therefore, this study evaluated the use of hyperspectral indices, red-edge parameters, and their combination to estimate and map the distributions of AGB and LAI for various growth stages of winter wheat. A hyperspectral sensor mounted on an unmanned aerial vehicle was used to obtain vegetation indices and red-edge parameters, and stepwise regression (SWR) and partial least squares regression (PLSR) methods were used to accurately estimate the AGB and LAI based on these vegetation indices, red-edge parameters, and their combination. The results show that: (i) most of the studied vegetation indices and red-edge parameters are significantly highly correlated with AGB and LAI; (ii) overall, the correlations between vegetation indices and AGB and LAI, respectively, are stronger than those between red-edge parameters and AGB and LAI, respectively; (iii) Compared with the estimations using only vegetation indices or red-edge parameters, the estimation of AGB and LAI using a combination of vegetation indices and red-edge parameters is more accurate; and (iv) The estimations of AGB and LAI obtained using the PLSR method are superior to those obtained using the SWR method. Therefore, combining vegetation indices with red-edge parameters and using the PLSR method can improve the estimation of AGB and LAI.

High-Throughput Prediction of Whole Season Green Area Index in Winter Wheat With an Airborne Multispectral Sensor

INTRODUCTION

In recent decades, the interest has grown to quantify the green area index as one of the key characteristics of crop canopies (e.g. for modelling transpiration, light interception, growth). The approach of estimating green area index based on multispectral reflection data from unmanned airborne vehicles with lightweight sensors might have the potential to deliver data with sufficient accuracy and high throughput during the whole season.

MATERIALS AND METHODS

We therefore examined the applicability of a recently launched drone-based multispectral system (Sequoia, Parrot) for the prediction of whole season green area index in winter wheat, with data from field trials in Northern Germany (2017, 2018 and 2019). The explanatory power of different modeling approaches to predict green area index based on multispectral data was tested: linear and non-linear regression models, multivariate techniques, and machine learning algorithms. Further, different predictors were implemented in these models: multispectral data as raw bands and as ratios. Additionally, a new approach for the evaluation of green area index predictions during senescence is introduced. It is shown that a robust calibration during growth phase is applicable during senescence as well.

RESULTS AND DISCUSSION

A linear model which includes all four wavebands provided by the sensor in three ratios (VIQUO) and a Support Vector Machine (SVM) algorithm allow a reliable and sufficiently accurate whole season prediction. The VIQUO-model is recommended as the best model, as it is precise but still relatively simple, thus easier to communicate and to apply than the SVM. The integrated values of predicted green area indices in an independent trial are highly correlated with their final biomass (R²: VIQUO = 0.84, SVM = 0.85) which represents the process of radiation interception, one of the determining factors of growths. This is an indicator for both, a robust model calibration and a high potential of the tested multispectral system for agricultural research and crop management.