Delayed spinetoram application is useful in managing Scirtothrips dorsalis Hood (Thysanoptera: Thripidae) in Florida strawberry

Scirtothrips dorsalis Hood (Thysanoptera: Thripidae) is an invasive, early-season pest of strawberry in Florida, causing feeding injury to young foliage that results in stunted plant growth and yield loss. Spinetoram, an effective insecticide for thrips pests with up to 3 applications per season permitted in strawberry, is often applied repeatedly during the early-season (Oct-Nov) to manage S. dorsalis, leaving few or no applications for flower thrips pests later in the season (Dec-Mar). Therefore, new strategies are needed to manage S. dorsalis with less insecticide, with the hypothesis that the first insecticide application can be delayed because young strawberry plants can compensate for minor feeding injury without compromising strawberry yield. Experiments conducted in strawberry field plots in Balm, FL, during 2018 and 2019 showed that delaying a spinetoram application for 14 days after infesting a plant with zero, 5, 10, or 20 S. dorsalis adults did not reduce the plant vigor and yield compared to spinetoram application after 4 days. Furthermore, young plants recovered from injury (10-30% bronzing injury on leaf veins and petioles) due to 1 or 2 S. dorsalis adults or larvae per trifoliate. A strategy of delaying the first spinetoram application when plants have 4-5 trifoliates should help reduce the number of insecticide applications needed for S. dorsalis management and reserve spinetoram applications for later in the season. Lower input costs in Florida strawberry without compromising yields due to thrips damage will improve the economics and sustainability of production systems.

Deep Learning for Strawberry Canopy Delineation and Biomass Prediction from High-Resolution Images

Modeling plant canopy biophysical parameters at the individual plant level remains a major challenge. This study presents a workflow for automatic strawberry canopy delineation and biomass prediction from high-resolution images using deep neural networks. High-resolution (5 mm) RGB orthoimages, near-infrared (NIR) orthoimages, and Digital Surface Models (DSM), which were generated by Structure from Motion (SfM), were utilized in this study. Mask R-CNN was applied to the orthoimages of two band combinations (RGB and RGB-NIR) to identify and delineate strawberry plant canopies. The average detection precision rate and recall rate were 97.28% and 99.71% for RGB images and 99.13% and 99.54% for RGB-NIR images, and the mean intersection over union (mIoU) rates for instance segmentation were 98.32% and 98.45% for RGB and RGB-NIR images, respectively. Based on the center of the canopy mask, we imported the cropped RGB, NIR, DSM, and mask images of individual plants to vanilla deep regression models to model canopy leaf area and dry biomass. Two networks (VGG-16 and ResNet-50) were used as the backbone architecture for feature map extraction. The R ² values of dry biomass models were about 0.76 and 0.79 for the VGG-16 and ResNet-50 networks, respectively. Similarly, the R ² values of leaf area were 0.82 and 0.84, respectively. The RMSE values were approximately 8.31 and 8.73 g for dry biomass analyzed using the VGG-16 and ResNet-50 networks, respectively. Leaf area RMSE was 0.05 m² for both networks. This work demonstrates the feasibility of deep learning networks in individual strawberry plant extraction and biomass estimation.

Prediction of Strawberry Dry Biomass from UAV Multispectral Imagery Using Multiple Machine Learning Methods

Biomass is a key biophysical parameter for precision agriculture and plant breeding. Fast, accurate and non-destructive monitoring of biomass enables various applications related to crop growth. In this paper, strawberry dry biomass weight was modeled using 4 canopy geometric parameters (area, average height, volume, standard deviation of height) and 25 spectral variables (5 band original reflectance values and 20 vegetation indices (VIs)) extracted from the Unmanned Aerial Vehicle (UAV) multispectral imagery. Six regression techniques—multiple linear regression (MLR), random forest (RF), support vector machine (SVM), multivariate adaptive regression splines (MARS), eXtreme Gradient Boosting (XGBoost) and artificial neural network (ANN)—were employed and evaluated for biomass prediction. The ANN had the highest accuracy in a five-fold cross-validation, with R2 of 0.89~0.93, RMSE of 7.16~8.98 g and MAE of 5.06~6.29 g. As for the other five models, the addition of VIs increased the R2 from 0.77~0.80 to 0.83~0.86, and reduced the RMSE from 8.89~9.58 to 7.35~8.09 g and the MAE from 6.30~6.70 to 5.25~5.47 g, respectively. Red-edge-related VIs, including the normalized difference red-edge index (NDRE), simple ratio vegetation index red-edge (SRRedEdge), modified simple ratio red-edge (MSRRedEdge) and chlorophyll index red and red-edge (CIred&RE), were the most influential VIs for biomass modeling. In conclusion, the combination of canopy geometric parameters and VIs obtained from the UAV imagery was effective for strawberry dry biomass estimation using machine learning models.

Combining canopy reflectance spectrometry and genome-wide prediction to increase response to selection for powdery mildew resistance in cultivated strawberry

High-throughput phenotyping is an emerging approach in plant science, but thus far only a few applications have been made in horticultural crop breeding. Remote sensing of leaf or canopy spectral reflectance can help breeders rapidly measure traits, increase selection accuracy, and thereby improve response to selection. In the present study, we evaluated the integration of spectral analysis of canopy reflectance and genomic information for the prediction of strawberry (Fragaria × ananassa) powdery mildew disease. Two multi-parental breeding populations of strawberry comprising a total of 340 and 464 pedigree-connected seedlings were evaluated in two separate seasons. A single-trait Bayesian prediction method using 1001 spectral wavebands in the ultraviolet-visible-near infrared region (350-1350 nm wavelength) combined with 8552 single nucleotide polymorphism markers showed up to 2-fold increase in predictive ability over models using markers alone. The integration of high-throughput phenotyping was further validated independently across years/trials with improved response to selection of up to 90%. We also conducted Bayesian multi-trait analysis using the estimated vegetative indices as secondary traits. Three vegetative indices (Datt3, REP_Li, and Vogelmann2) had high genetic correlations (rA) with powdery mildew visual ratings with average rA values of 0.76, 0.71, and 0.71, respectively. Increasing training population sizes by incorporating individuals with only vegetative index information yielded substantial increases in predictive ability. These results strongly indicate the use of vegetative indices as secondary traits for indirect selection. Overall, combining spectrometry and genome-wide prediction improved selection accuracy and response to selection for powdery mildew resistance, demonstrating the power of an integrated phenomics-genomics approach in strawberry breeding.

Beyond Hydrography - Marine Geomatics at the University of Florida

Florida depends on the oceans, yet its waters have not been extensively mapped to the highest standards. While there is a need for marine spatial data for a wide range of applications and issues, there is also a need to develop data acquisition, processing, and analytical workflows and to integrate different surveying instruments that can capture the complex and extensive coastal environment – both above and below the waterline. This note provides an overview of the research performed by scientists at the School of Forest, Fisheries, and Geomatics Sciences, University of Florida, in the field of hydrography and marine geomatics.

Bathymetry inversion with optimal Sentinel-2 imagery using random forest modeling.. [DOI: 10.5194/egusphere-egu22-10426] [Citation(s) in RCA: 0] [Impact Index Per Article: 0]

<p>Bathymetry inversion using remote sensing techniques is a topic of increasing interest in coastal management and monitoring. Freely accessible Sentinel-2 imagery offers high-resolution multispectral data that enables bathymetry inversion in optically shallow waters. This study presents a framework leading to a generalized Satellite-Derived Bathymetry (SDB) model applicable to vast and diversified coastal regions utilizing multi-date images. A multivariate regression random forest model was used to derive bathymetry from optimal Sentinel-2 images over an extensive 210 km coastal stretch along southwestern Florida (United States). Model calibration and validation were done using airborne lidar bathymetry (ALB) data. As ALB surveys are costly, the proposed model was trained with a limited and practically feasible ALB data sample to expand the model’s practicality. Using multi-image bands as individual features in the random forest model yielded high accuracy with root-mean-square error values of 0.42 m and lower for depths up to 13 m.</p>

Carbon and nitrogen pools and mobile fractions in surface soils across a mangrove saltmarsh ecotone

In the subtropics, climate change is pushing woody mangrove forests into herbaceous saltmarshes, altering soil carbon (C) and nitrogen (N) pools, with implications for coastal wetland productivity and C and N exports. We quantified total C and N pools, and mobile fractions including extractable mineral N, extractable organic C and N, and active (aerobically mineralizable) C and N, in surface soils (top 7.6 cm) of adjacent mangrove (primarily Avicennia germinans) and saltmarsh (Juncus roemerianus) vegetation zones in tidal wetlands of west-central Florida (USA). We tested whether surface-soil accumulations of C, N, and their potentially mobile fractions are greater in mangrove than in saltmarsh owing to greater accumulations in the mangrove zone of soil organic matter (SOM) and fine mineral particles (C- and N-retaining soil constituents). Extractable organic fractions were 39-45% more concentrated in mangrove than in saltmarsh surface soil, and they scaled steeply and positively with SOM and fine mineral particle (silt + clay) concentrations, which themselves were likewise greater in mangrove soil. Elevation may drive this linkage. Mangrove locations were generally at lower elevations, which tended to have greater fine particle content in the surface soil. Active C and extractable mineral N were marginally (p < 0.1) greater in mangrove soil, while active N, total N, and total C showed no statistical differences between zones. Extractable organic C and N fractions composed greater shares of total C and N pools in mangrove than in saltmarsh surface soils, which is meaningful for ecosystem function, as persistent leaching of this fraction can perpetuate nutrient limitation. The active (mineralizable) C and N fractions we observed constituted a relatively small component of total C and N pools, suggesting that mangrove surface soils may export less C and N than would be expected from their large total C and N pools.

Deep Learning Classification of High-Resolution Drone Images Using the ArcGIS Pro Software

Deep learning classification of invasive species using widely-used ArcGIS Pro software and increasingly common drone imagery can aid in identification and management of natural areas. A step-by-step implementation, with associated data for users to access, is presented to make this technology more widely accessible to GIS analysts, researchers, and graduate students working with remotely sensed data in the natural resource field.

A Step-by-Step Guide for Automated Plant Canopy Delineation Using Deep Learning: An Example in Strawberry Using ArcGIS Pro Software

This publication presents a guide to image analysis for researchers and farm managers who use ArcGIS software. Anyone with basic geographic information system analysis skills may follow along with the demonstration and learn to implement the Mask Region Convolutional Neural Networks model, a widely used model for object detection, to delineate strawberry canopies using ArcGIS Pro Image Analyst Extension in a simple workflow. This process is useful for precision agriculture management.

Eco-physiological evaluation of multipurpose tree species to ameliorate saline soils

Salinity is a widespread soil and underground water contaminant threatening food security and economic stability. Phytoremediation is an efficient and environmental-friendly solution to mitigate salinity impacts. The present study was conducted to evaluate the phytoremediation potential of five multipurpose trees: Vachellia nilotica, Concorpus erectus, Syzygium cumini, Tamarix aphylla and Eucalyptus cammaldulensis under four salinity treatments: Control, 10, 20 and 30 dS m^-1. Salinity negatively impacted all the tested species. However, E. cammaldulensis and T. aphylla exhibited the lowest reduction (28%) and (35%) in plant height respectively along with a minimal reduction in leaf gas exchange while V. nilotica, S. cumini and C. erectus showed severe dieback. Similarly, the antioxidant enzymes increased significantly in E. cammaldulensis and T. aphylla as Superoxide Dismutase (87% and 79%), Catalase (66% and 67%) and Peroxidase (89% and 81%), respectively. Furthermore, both of these species maintained optimum Na/K ratio reducing the highest levels of soil EC_e and SAR, suggesting the best phytoremediation potential. The present study identifies that E. cammaldulensis and T. aphylla showed effective tolerance mechanisms and the highest salt sequestration; therefore, may be used for phyto-amelioration of salinity impacted lands. Novelty statement Although previous studies evaluated the tolerance potential of many tree species, comparative and physiochemical evaluation of multipurpose tree species has been remained unexplored. In this scenario, eco-physiological characterization of multipurpose tree species may inform tree species for phytoremediation of saline soils according to the level of salinity. Optimizing tree species selection also improves the success of wood for energy and revenue generation while restoring degraded soils.

Scan Pattern Characterization of Velodyne VLP-16 Lidar Sensor for UAS Laser Scanning

Many lightweight lidar sensors employed for UAS lidar mapping feature a fan-style laser emitter-detector configuration which results in a non-uniform pattern of laser pulse returns. As the role of UAS lidar mapping grows in both research and industry, it is imperative to understand the behavior of the fan-style lidar sensor to ensure proper mission planning. This study introduces sensor modeling software for scanning simulation and analytical equations developed in-house to characterize the non-uniform return density (i.e., scan pattern) of the fan-style sensor, with special focus given to a popular fan-style sensor, the Velodyne VLP-16 laser scanner. The results indicate that, despite the high pulse frequency of modern scanners, areas of poor laser pulse coverage are often present along the scanning path under typical mission parameters. These areas of poor coverage appear in a variety of shapes and sizes which do not necessarily correspond to the forward speed of the scanner or the height of the scanner above the ground, highlighting the importance of scan simulation for proper mission planning when using a fan-style sensor.

Florida’s Urban Forest: A Valuation of Benefits

This new 13-page article combines canopy coverage data from all of Florida's metropolitan and micropolitan areas with ecological models developed by the USDA Forest Service to calculate several key benefits of urban trees and an approximation of their monetary value. Benefits of urban trees include carbon sequestration/storage, air pollution filtration, and stormwater mitigation. Written by Drew C. McLean, Andrew K. Koeser, Deborah R. Hilbert, Shawn Landry, Amr Abd-Elrahman, Katie Britt, Mary Lusk, Michael G. Andreu, and Robert J. Northrop, and published by the UF/IFAS Environmental Horticulture Department.https://edis.ifas.ufl.edu/ep595

Evaluating the effect of remote sensing image spatial resolution on soil exchangeable potassium prediction models in smallholder farm settings

Major end users of Digital Soil Mapping (DSM) such as policy makers and agricultural extension workers are faced with choosing the appropriate remote sensing data. The objective of this research is to analyze the spatial resolution effects of different remote sensing images on soil prediction models in two smallholder farms in Southern India called Kothapally (Telangana State), and Masuti (Karnataka State), and provide empirical guidelines to choose the appropriate remote sensing images in DSM. Bayesian kriging (BK) was utilized to characterize the spatial pattern of exchangeable potassium (K_ex) in the topsoil (0-15 cm) at different spatial resolutions by incorporating spectral indices from Landsat 8 (30 m), RapidEye (5 m), and WorldView-2/GeoEye-1/Pleiades-1A images (2 m). Some spectral indices such as band reflectances, band ratios, Crust Index and Atmospherically Resistant Vegetation Index from multiple images showed relatively strong correlations with soil K_ex in two study areas. The research also suggested that fine spatial resolution WorldView-2/GeoEye-1/Pleiades-1A-based and RapidEye-based soil prediction models would not necessarily have higher prediction performance than coarse spatial resolution Landsat 8-based soil prediction models. The end users of DSM in smallholder farm settings need select the appropriate spectral indices and consider different factors such as the spatial resolution, band width, spectral resolution, temporal frequency, cost, and processing time of different remote sensing images. Overall, remote sensing-based Digital Soil Mapping has potential to be promoted to smallholder farm settings all over the world and help smallholder farmers implement sustainable and field-specific soil nutrient management scheme.

Spatial downscaling of soil prediction models based on weighted generalized additive models in smallholder farm settings

Digital soil mapping (DSM) is gaining momentum as a technique to help smallholder farmers secure soil security and food security in developing regions. However, communications of the digital soil mapping information between diverse audiences become problematic due to the inconsistent scale of DSM information. Spatial downscaling can make use of accessible soil information at relatively coarse spatial resolution to provide valuable soil information at relatively fine spatial resolution. The objective of this research was to disaggregate the coarse spatial resolution soil exchangeable potassium (K_ex) and soil total nitrogen (TN) base map into fine spatial resolution soil downscaled map using weighted generalized additive models (GAMs) in two smallholder villages in South India. By incorporating fine spatial resolution spectral indices in the downscaling process, the soil downscaled maps not only conserve the spatial information of coarse spatial resolution soil maps but also depict the spatial details of soil properties at fine spatial resolution. The results of this study demonstrated difference between the fine spatial resolution downscaled maps and fine spatial resolution base maps is smaller than the difference between coarse spatial resolution base maps and fine spatial resolution base maps. The appropriate and economical strategy to promote the DSM technique in smallholder farms is to develop the relatively coarse spatial resolution soil prediction maps or utilize available coarse spatial resolution soil maps at the regional scale and to disaggregate these maps to the fine spatial resolution downscaled soil maps at farm scale.

Spatially-explicit modeling of multi-scale drivers of aboveground forest biomass and water yield in watersheds of the Southeastern United States

Understanding ecosystem processes and the influence of regional scale drivers can provide useful information for managing forest ecosystems. Examining more local scale drivers of forest biomass and water yield can also provide insights for identifying and better understanding the effects of climate change and management on forests. We used diverse multi-scale datasets, functional models and Geographically Weighted Regression (GWR) to model ecosystem processes at the watershed scale and to interpret the influence of ecological drivers across the Southeastern United States (SE US). Aboveground forest biomass (AGB) was determined from available geospatial datasets and water yield was estimated using the Water Supply and Stress Index (WaSSI) model at the watershed level. Our geostatistical model examined the spatial variation in these relationships between ecosystem processes, climate, biophysical, and forest management variables at the watershed level across the SE US. Ecological and management drivers at the watershed level were analyzed locally to identify whether drivers contribute positively or negatively to aboveground forest biomass and water yield ecosystem processes and thus identifying potential synergies and tradeoffs across the SE US region. Although AGB and water yield drivers varied geographically across the study area, they were generally significantly influenced by climate (rainfall and temperature), land-cover factor1 (Water and barren), land-cover factor2 (wetland and forest), organic matter content high, rock depth, available water content, stand age, elevation, and LAI drivers. These drivers were positively or negatively associated with biomass or water yield which significantly contributes to ecosystem interactions or tradeoff/synergies. Our study introduced a spatially-explicit modelling framework to analyze the effect of ecosystem drivers on forest ecosystem structure, function and provision of services. This integrated model approach facilitates multi-scale analyses of drivers and interactions at the local to regional scale.

Classifying spatially heterogeneous wetland communities using machine learning algorithms and spectral and textural features

Mapping of wetlands (marsh vs. swamp vs. upland) is a common remote sensing application.Yet, discriminating between similar freshwater communities such as graminoid/sedge fromremotely sensed imagery is more difficult. Most of this activity has been performed using medium to low resolution imagery. There are only a few studies using highspatial resolutionimagery and machine learning image classification algorithms for mapping heterogeneouswetland plantcommunities. This study addresses this void by analyzing whether machine learning classifierssuch as decisiontrees (DT) and artificial neural networks (ANN) can accurately classify graminoid/sedgecommunities usinghigh resolution aerial imagery and image texture data in the Everglades National Park, Florida.In addition tospectral bands, the normalized difference vegetation index, and first- and second-order texturefeatures derivedfrom the near-infrared band were analyzed. Classifier accuracies were assessed using confusiontablesand the calculated kappa coefficients of the resulting maps. The results indicated that an ANN(multilayerperceptron based on backpropagation) algorithm produced a statistically significantly higheraccuracy(82.04%) than the DT (QUEST) algorithm (80.48%) or the maximum likelihood (80.56%)classifier (α<0.05). Findings show that using multiple window sizes provided the best results. First-ordertexture featuresalso provided computational advantages and results that were not significantly different fromthose usingsecond-order texture features.

Classification of Submerged Aquatic Vegetation in Black River Using Hyperspectral Image Analysis

Monitoring aquatic vegetation is an important component of water resource management due to the ecological services provided by these habitats. Spectrally-rich hyperspectral imagery can be an efficient tool for mapping and classifying macrophyte communities. Identification of submerged vegetation in aquatic regions is complicated by variations in optical properties of water constituents, sun-water-sensor geometry, water depth and the spectral/structural complexity of the plants. Many studies have attempted to detect aquatic vegetation in coastal waters; however, few studies have targeted shallow, black-water rivers tainted with chromophoric dissolved organic matter (CDOM). This study investigates methods to analyze airborne hyperspectral imagery and detect and classify aquatic vegetation in a black-water riverine system. Images were normalized to account for reflectance from the water surface and varying water depth before being analyzed by the Maximum Likelihood (ML) and three other non-parametric classifiers: Artificial Neural Network (ANN), Support Vector Machine (SVM) and Spectral Angular Mapper (SAM). Quality assessment analysis indicated a general classification and detection accuracy improvement when non-parametric classifiers were applied on the normalized and depth invariant images. A maximum classification accuracy of about 69% was achieved when the ANN classifier was applied on the normalized images, and maximum detection accuracies of 93% and 92% were obtained when the SAM and the SVM classifiers were applied on depth invariant images, respectively.

Mapping potential carbon and timber losses from hurricanes using a decision tree and ecosystem services driver model

Information on the effect of direct drivers such as hurricanes on ecosystem services is relevant to landowners and policy makers due to predicted effects from climate change. We identified forest damage risk zones due to hurricanes and estimated the potential loss of 2 key ecosystem services: aboveground carbon storage and timber volume. Using land cover, plot-level forest inventory data, the Integrated Valuation of Ecosystem Services and Tradeoffs (InVEST) model, and a decision tree-based framework; we determined potential damage to subtropical forests from hurricanes in the Lower Suwannee River (LS) and Pensacola Bay (PB) watersheds in Florida, US. We used biophysical factors identified in previous studies as being influential in forest damage in our decision tree and hurricane wind risk maps. Results show that 31% and 0.5% of the total aboveground carbon storage in the LS and PB, respectively was located in high forest damage risk (HR) zones. Overall 15% and 0.7% of the total timber net volume in the LS and PB, respectively, was in HR zones. This model can also be used for identifying timber salvage areas, developing ecosystem service provision and management scenarios, and assessing the effect of other drivers on ecosystem services and goods.

A framework for identifying carbon hotspots and forest management drivers

Spatial analyses of ecosystem system services that are directly relevant to both forest management decision making and conservation in the subtropics are rare. Also, frameworks that identify and map carbon stocks and corresponding forest management drivers using available regional, national, and international-level forest inventory datasets could provide insights into key forest structural characteristics and management practices that are optimal for carbon storage. To address this need we used publicly available USDA Forest Service Forest Inventory and Analysis data and spatial analyses to develop a framework for mapping "carbon hotspots" (i.e. areas of significantly high tree and understory aboveground carbon stocks) across a range of forest types using the state of Florida, USA as an example. We also analyzed influential forest management variables (e.g. forest types, fire, hurricanes, tenure, management activities) using generalized linear mixed modeling to identify drivers associated with these hotspots. Most of the hotspots were located in the northern third of the state some in peri-urban areas, and there were no identifiable hotspots in South Florida. Forest silvicultural treatments (e.g. site preparation, thinning, logging, etc) were not significant predictors of hotspots. Forest types, site quality, and stand age were however significant predictors. Higher site quality and stand age increased the probability of forests being classified as a hotspot. Disturbance type and time since disturbance were not significant predictors in our analyses. This framework can use globally available forest inventory datasets to analyze and map ecosystems service provision areas and bioenergy supplies and identify forest management practices that optimize these services in forests.