Leaf Area Index Inversion of Spartina alterniflora Using UAV Hyperspectral Data Based on Multiple Optimized Machine Learning Algorithms

We compared direct and indirect estimates of leaf area index (LAI) for lodgepole and loblolly pine stands. Indirect estimates of LAI using radiative methods of the LI-COR LAI-2000 Plant Canopy Analyzer (PCA) did not correlate with allometric estimates for lodgepole pine, and correlated only weakly with litter-trap estimates for loblolly pine. The PCA consistently under-estimated LAI in lodgepole pine stands with high LAI, and over-estimated LAI in the loblolly pine stands with low LAI. We developed a physical model to test the hypothesis that the PCA may under-estimate LAI in high leaf area stands because of increased foliage overlap and, therefore, increased selfshading. Radiative estimates of LAI using the PCA for the physical model were consistenly lower than allometric measures. Results from the physical model suggested that increased foliage overlap decreased the ability of the PCA to accurately estimate LAI. The relationship between allometric and radiative measures suggested an upper asymptote in LAI estimated using the PCA. The PCA may not accurately estimate LAI in stands of low or high leaf area index, and the bias or error associated with these estimates probably depends on species and canopy structure. A species specific correction factor will not necessarily correct bias in LAI estimates using the PCA.

Precise and timely leaf area index (LAI) estimation for winter wheat is crucial for precision agriculture. The emergence of high-resolution unmanned aerial vehicle (UAV) data and machine learning techniques offers a revolutionary approach for fine-scale estimation of wheat LAI at the low cost. While machine learning has proven valuable for LAI estimation, there are still model limitations and variations that impede accurate and efficient LAI inversion. This study explores the potential of classical machine learning models and deep learning model for estimating winter wheat LAI using multispectral images acquired by drones. Initially, the texture features and vegetation indices served as inputs for the partial least squares regression (PLSR) model and random forest (RF) model. Then, the ground-measured LAI data were combined to invert winter wheat LAI. In contrast, this study also employed a convolutional neural network (CNN) model that solely utilizes the cropped original image for LAI estimation. The results show that vegetation indices outperform the texture features in terms of correlation analysis with LAI and estimation accuracy. However, the highest accuracy is achieved by combining both vegetation indices and texture features to invert LAI in both conventional machine learning methods. Among the three models, the CNN approach yielded the highest LAI estimation accuracy (R 2 = 0.83), followed by the RF model (R 2 = 0.82), with the PLSR model exhibited the lowest accuracy (R 2 = 0.78). The spatial distribution and values of the estimated results for the RF and CNN models are similar, whereas the PLSR model differs significantly from the first two models. This study achieves rapid and accurate winter wheat LAI estimation using classical machine learning and deep learning methods. The findings can serve as a reference for real-time wheat growth monitoring and field management practices.

The leaf area index (LAI) and the clumping index (CI) provide valuable insight into the spatial patterns of forest canopies, the canopy light regime and forest productivity. This study examines the spatial patterns of LAI and CI in a boreal mixed-wood forest, using extensive field measurements and remote sensing analysis. The objectives of this study are to: (1) examine the utility of airborne lidar (light detection and ranging) and hyperspectral data to model LAI and clumping indices; (2) compare these results to those found from commonly used Landsat vegetation indices (i.e. the normalized difference vegetation index (NDVI) and the simple ratio (SR)); (3) determine whether the fusion of lidar data with Landsat and/or hyperspectral data will improve the ability to model clumping and LAI; and (4) assess the relationships between clumping, LAI and canopy biochemistry. Regression models to predict CI were much stronger than those for LAI at the site. Lidar was the single best predictor of CI (r 2 > 0.8). Landsat NDVI and SR also had a moderately strong predictive performance for CI (r 2 > 0.68 with simple linear and non-linear regression forms), suggesting that canopy clumping can be predicted operationally from satellite platforms, at least in boreal mixed-wood environments. Foliar biochemistry, specifically canopy chlorophyll, carotenoids, magnesium, phosphorus and nitrogen, was strongly related to the clumping index. Combined, these results suggest that Landsat models of clumping could provide insight into the spatial distribution of foliar biochemistry, and thereby photosynthetic capacity, for boreal mixed-wood canopies. LAI models were weak (r 2 < 0.4) unless separate models were used for deciduous and coniferous plots. Coniferous LAI was easier to model than deciduous LAI (r 2 > 0.8 for several indices). Deciduous models of LAI were weaker for all remote sensing indices (r 2 < 0.67). There was a strong, linear relationship between foliar biochemistry and LAI for the deciduous plots. Overall, our results suggest that broadband satellite indices have strong predictive performance for clumping, but that airborne hyperspectral or lidar data are required to develop strong models of LAI at this boreal mixed-wood site.

Leaf area index (LAI) is a key ecological indicator for describing the structure of canopies and for modelling energy exchange between atmosphere and biosphere. While LAI of the forest overstory can be accurately assessed over large spatial scales via remote sensing, LAI of the forest understory (LAIu) is still largely ignored in ecological studies and ecosystem modelling due to the fact that it is often too complex to be destructively sampled or approximated by other site parameters. Additionally, so far only few attempts have been made to retrieve understory LAI via remote sensing, because dense canopies with high LAI are often hindering retrieval algorithms to produce meaningful estimates for understory LAI. Consequently, the forest understory still constitutes a poorly investigated research realm impeding ecological studies to properly account for its contribution to the energy absorption capacity of forest stands. This study aims to compare three conceptually different indirect retrieval methodologies for LAIu over a diverse panel of forest understory types distributed across Europe. For this we carried out near-to-surface measurements of understory reflectance spectra as well as digital surface photography over the extended network of Integrated Carbon Observation System (ICOS) forest ecosystem sites. LAIu was assessed by exploiting the empirical relationship between vegetation cover and light absorption (Beer-Lambert- Bouguer law) as well as by utilizing proposed relationships with two prominent vegetation indices: normalized difference vegetation index (NDVI) and simple ratio (SR). Retrievals from the three methods were significantly correlated with each other (r = 0.63–0.99, RMSE = 0.53–0.72), but exhibited also significant bias depending on the LAI scale. The NDVI based retrieval approach most likely overestimates LAI at productive sites when LAIu > 2, while the simple ratio algorithm overestimates LAIu at sites with sparse understory vegetation and presence of litter or bare soil. The purely empirical method based on the Beer-Lambert law of light absorption seems to offer a good compromise, since it provides reasonable LAIu values at both low and higher LAI ranges. Surprisingly, LAIu variation among sites seems to be largely decoupled from differences in climate and light permeability of the overstory, but significantly increased with vegetation diversity (expressed as species richness) and hence proposes new applications of LAIu in ecological modelling.

Effects of maize and cassava on the performance of intercropped egusi melon ( Citrullus lanatus (L.) Thunb.) and okra ( Abelmoschus esculentus (L.) Moench.) in Nigeria

Rapid and accurate crop chlorophyll content estimation and the leaf area index (LAI) are both crucial for guiding field management and improving crop yields. This paper proposes an accurate monitoring method for LAI and soil plant analytical development (SPAD) values (which are closely related to leaf chlorophyll content; we use the SPAD instead of chlorophyll relative content) based on the fusion of ground–air multi-source data. Firstly, in 2020 and 2021, we collected unmanned aerial vehicle (UAV) multispectral data, ground hyperspectral data, UAV visible-light data, and environmental cumulative temperature data for multiple growth stages of summer maize, respectively. Secondly, the effective plant height (canopy height model (CHM)), effective accumulation temperature (growing degree days (GDD)), canopy vegetation index (mainly spectral vegetation index) and canopy hyperspectral features of maize were extracted, and sensitive features were screened by correlation analysis. Then, based on single-source and multi-source data, multiple linear regression (MLR), partial least-squares regression (PLSR) and random forest (RF) regression were used to construct LAI and SPAD inversion models. Finally, the distribution of LAI and SPAD prescription plots was generated and the trend for the two was analyzed. The results were as follows: (1) The correlations between the position of the hyperspectral red edge and the first-order differential value in the red edge with LAI and SPAD were all greater than 0.5. The correlation between the vegetation index, including a red and near-infrared band, with LAI and SPAD was above 0.75. The correlation between crop height and effective accumulated temperature with LAI and SPAD was above 0.7. (2) The inversion models based on multi-source data were more effective than the models made with single-source data. The RF model with multi-source data fusion achieved the highest accuracy of all models. In the testing set, the LAI and SPAD models’ R2 was 0.9315 and 0.7767; the RMSE was 0.4895 and 2.8387. (3) The absolute error between the extraction result of each model prescription map and the measured value was small. The error between the predicted value and the measured value of the LAI prescription map generated by the RF model was less than 0.4895. The difference between the predicted value and the measured value of the SPAD prescription map was less than 2.8387. The LAI and SPAD of summer maize first increased and then decreased with the advancement of the growth period, which was in line with the actual growth conditions. The research results indicate that the proposed method could effectively monitor maize growth parameters and provide a scientific basis for summer maize field management.

Accurate leaf area index estimation in sorghum using high-resolution UAV data and machine learning models

Inversion of the PROSAIL model to estimate leaf area index of maize, potato, and sunflower fields from unmanned aerial vehicle hyperspectral data

The use of unmanned aerial vehicles (UAVs) is becoming increasingly widespread in many areas of human life every year. Not only military, but also civilian use of UAVs in various fields of science, industry, forestry, agronomy, cartography, ecology, and space research is hard to overestimate. For this reason, the use and development of UAVs in various fields remains extremely promising in the near future. Compared to manned vehicles, UAVs have the following advantages: no crew and no need for life support systems, large airfields; lower cost and low expenses for their development, production and operation; better weight and dimensions; high reliability and maneuverability; a wide range of specialized equipment for placement on board. UAVs are being introduced into all spheres of human life, even those where UAVs have no manned alternative. Some models of multicopters are already being used to deliver goods and extinguish fires. UAVs of new designs continue to be developed.In all areas of UAV application, the task of increasing autonomy remains relevant and important. The duration of con-tinuous operation and flight range directly affect the capabilities of the UAV, its ability to perform tasks and efficiency.UAV power consumption depends on the weight of the payload, weather conditions and drone speed. Energy consumption is a critical criterion for a drone to perform its missions and needs to be further studied. Many studies have been conducted to increase flight time by improving energy efficiency or reducing the energy consumed by the drone.Study the influence of battery parameters (capacity and weight) and propeller group (number of propellers and their diameter) on the maximum possible flight duration of UAVs of different masses.The main factors affecting the maximum flight duration of a UAV are considered. A model for calculating the power consumed by the device in the hover mode is presented. The maximum flight duration of a UAV is calculated depending on the battery capacity and weight, the number of propellers, and the diameter of the propellers for vehicles of different masses.It is shown that there is no point in increasing the battery capacity indefinitely, and there is an optimum point that depends on the ratio of the battery mass to the mass of the vehicle. It is also shown that increasing the diameter of the propellers significantly increases the flight time, while increasing their number is not as effective.

Leaf Area Index (LAI) and biomass (BIO) are essential agronomic parameters that reflect the growth of potatoes and are related to their biomass. Their precise estimation is capable of monitoring crop growth, guiding field management, and optimizing planting spatial patterns. Traditional potato leaf area indexing and biomass estimation primarily rely on field sampling surveys. This method is low in efficiency, high in cost, and limited by the number of samples. It cannot accurately reflect potato growth and meet the real-time estimation needs of large areas. Compared to the use of satellite remote sensing data (RSD) for estimating LAI and biomass, research on estimating these two phenotypic parameters using crewless aerial vehicle (UAV) RSD is relatively immature. Research on estimating crop growth index parameters by remote sensing primarily focuses on data obtained from specific types of sensors, targeting specific growth stages to compare and analyze the accuracy of different methods. However, there are few estimates of the impact of optimizing the best data types and optimal growth stage for LAI and biomass estimation by comparing and analyzing different sensor data and different growth stages. Multi-sensor integration technology has made it possible to study different crop phenotype information and estimate the best data type and optimal growth stage in crop phenotypic data estimation, establishing it as a new hot spot in the field. This paper integrates high-definition digital cameras and imaging hyperspectrometers on the UAV platform to obtain digital images and hyperspectral data simultaneously, along with ground-measured potato leaf area index and biomass data. Using the partial least squares regression (PLSR), random forest (RF), support vector machine (SVM), and backpropagation (BP) neural network methods, we got digital images and hyperspectral data from different stages of growth, put together a digital image index and a vegetation index, and looked at how they related to LAI and BIO. Then, we chose the index that had the strongest correlation. To establish LAI and biomass estimation models at various growth phases, this paper compared and analyzed the estimation impacts of various data types and models at various growth phases. It then selected the best data types for LAI estimation and biomass estimation at different growth stages, as well as the best growth phases for LAI and biomass estimation. The outcomes indicated that when potato LAI was estimated, the mean values of R 2 and RMSE of the four estimation models were 0.75 and 0.30 Kg/mu at the tuber growth stage, respectively, and the estimation effect was the best, indicating that this was the best growth phase for LAI estimation. The average values of R 2 and RMSE in the LAI estimation model using the hyperspectral vegetation index were 0.73 and 0.33 Kg/mu, respectively, indicating that hyperspectral data was the best data type for LAI estimation. When potato biomass was estimated, the mean values of R 2 and RMSE of the four methods were 0.67 and 15.25 Kg/mu, respectively, at the tuber growth stage, which were better than other growth phases, demonstrating that this was the best growth phase for biomass estimation. The average values of R 2 and RMSE of the biomass estimation model using the hyperspectral vegetation index were 0.67 and 20.08 Kg/mu, respectively, indicating that the hyperspectral data was the best data type for biomass estimation. The average values of R 2 of the LAI and biomass estimation model at the maturity stage were only 0.56 and 0.36, both of which indicated poor estimation effects. Our study can serve as a guide to selecting the most effective method for estimating parameters for essential indexes in crop growth monitoring.

Leaf pigment contents, such as chlorophylls a and b content (C a b ) or carotenoid content (Car), and the leaf area index (LAI) are recognized indicators of plants’ and forests’ health status that can be estimated through hyperspectral imagery. Their measurement on a seasonal and yearly basis is critical to monitor plant response and adaptation to stress, such as droughts. While extensively done over dense canopies, estimation of these variables over tree-grass ecosystems with very low overstory LAI (mean site LAI &lt; 1 m 2 /m 2 ), such as woodland savannas, is lacking. We investigated the use of look-up table (LUT)-based inversion of a radiative transfer model to retrieve LAI and leaf C a b and Car from AVIRIS images at an 18 m spatial resolution at multiple dates over a broadleaved woodland savanna during the California drought. We compared the performances of different cost functions in the inversion step. We demonstrated the spatial consistency of our LAI, C a b , and Car estimations using validation data from low and high canopy cover parts of the site, and their temporal consistency by qualitatively confronting their variations over two years with those that would be expected. We concluded that LUT-based inversions of medium-resolution hyperspectral images, achieved with a simple geometric representation of the canopy within a 3D radiative transfer model (RTM), are a valid means of monitoring woodland savannas and more generally sparse forests, although for maximum applicability, the inversion cost functions should be selected using validation data from multiple dates. Validation revealed that for monitoring use: The normalized difference vegetation index (NDVI) outperformed other indices for LAI estimations (root mean square error (RMSE) = 0.22 m 2 /m 2 , R 2 = 0.81); the band ratio ρ 0.750 μ m ρ 0.550 μ m retrieved C a b more accurately than other chlorophylls indices (RMSE = 5.21 μ g/cm 2 , R 2 = 0.73); RMSE over the 0.5–0.55 μ m interval showed encouraging results for Car estimations.

Abstract. The rapid development of unmanned aerial vehicles is a challenge for applied research. Many technologies are developed and then researcher are looking up for their application in different sectors. Therefore, we decided to verify the use of the unmanned aerial vehicle for wood chips pile monitoring. We compared the use of GNSS device and unmanned aerial vehicle for volume estimation of four wood chips piles. We used DJI Phantom 3 Professional with the built-in camera and GNSS device (geoexplorer 6000). We used Agisoft photoscan for processing photos and ArcGIS for processing points. Volumes calculated from pictures were not statistically significantly different from amounts calculated from GNSS data and high correlation between them was found (p = 0.9993). We conclude that the use of unmanned aerial vehicle instead of the GNSS device does not lead to significantly different results. Tthe data collection consumed from almost 12 to 20 times less time with the use of UAV. Additionally, UAV provides documentation trough orthomosaic.

Controls on stand transpiration and soil water utilization along a tree density gradient in a Neotropical savanna

Automated Papaya Farm Monitoring system using Unmanned Aerial Vehicle (UAV)

Effect of shade regimes on gradients of microclimate, canopy extent (leaf area index: LAI) and light integrals in dry season field-grown cocoa (cacao) seedlings was investigated in a rainforest zone of Nigeria. The shade regimes tested were: unshaded/open-to-sun, dense shade and moderate shade. Shade intensity affected solar radiation transmission through cacao canopy, photosynthetic active radiation (PAR) and canopy light attenuation (extinction coefficient, k). Intensity of transmitted radiation below the canopy from incident radiation was highest for open-to-sun, followed by moderate and dense shade, respectively. The temporal trend of intercepted radiation showed that intercepted radiation increased from December to May, and, the values were highest for open-to-sun, followed by moderate and dense shade. The ratio of transmitted (Io) to incident (I) radiation (IO/I) was higher for open sun. Significant differences were found between open-to-sun (unshaded) and moderate and dense shade intensity for value of canopy extinction coefficient (k). The association of growing degree days (GDD), and, total leaf number (TLN) and leaf area index (LAI), were characterized by high coefficient of determination (R2) for the respective open, dense and moderate shade treatments. Inverse of the slope of the regression of relationship between estimated thermal time (°Cdays) and corresponding total leaf number (TLN) denotes leaf appearance rate (phyllochron, in °Cdays/leaf). Rate of leaf appearance was faster in open sun compared with to that in moderate or dense shade intensity. Characteristics of the cacao canopy development were measured by leaf area index (LAI), a parameter which affects the intercepted photosynthetic active radiation (PAR). Higher LAI was obtained in no shade (open sun) compared to that in moderate or dense shade treatments. Unshaded plants had a higher radiation use efficiency (RUE) and RUE values were significantly higher compared to the other two treatments. Low light intensity and LAI for under-storey cacao had negative implications for growth and biomass development. Air temperatures within the cacao field were highest for open sun cacao, followed by moderate and dense shade, respectively; the values increased from December to April, with peak values seen in April.