Blue Band Reflectance Research Articles

Impact Assessment of Nematode Infestation on Soybean Crop Production Using Aerial Multispectral Imagery and Machine Learning

Accurate and prompt estimation of geospatial soybean yield (SY) is critical for the producers to determine key factors influencing crop growth for improved precision management decisions. This study aims to quantify the impacts of soybean cyst nematode (SCN) infestation on soybean production and the yield of susceptible and resistant seed varieties. Susceptible varieties showed lower yield and crop vigor recovery, and high SCN population (20 to 1080) compared to resistant varieties (SCN populations: 0 to 340). High-resolution (1.3 cm/pixel) aerial multispectral imagery showed the blue band reflectance (r = 0.58) and Green Normalized Difference Vegetation Index (GNDVI, r = −0.6) have the best correlation with the SCN populations. While GDNVI, Green Chlorophyll Index (GCI), and Normalized Difference Red Edge Index (NDRE) were the best differentiators of plant vigor and had the highest correlation with SY (r = 0.59–0.75). Reflectance (REF) and VIs were then used for SY estimation using two statistical and four machine learning (ML) models at 10 different train–test data split ratios (50:50–95:5). The ML models and train–test data split ratio had significant impacts on SY estimation accuracy. Random forest (RF) was the best and consistently performing model (r: 0.84–0.97, rRMSE: 8.72–20%), while a higher train–test split ratio lowered the performances of the ML models. The 95:5 train–test ratio showed the best performance across all the models, which may be a suitable ratio for modeling over smaller or medium-sized datasets. Such insights derived using high spatial resolution data can be utilized to implement precision crop protective operations for enhanced soybean yield and productivity.

Spatial mapping of soil moisture content using very-high resolution UAV-based multispectral image analytics

Assessing soil moisture content (SMC) is necessary for managing water at a spatial scale. Remote sensing technologies provide a robust approach for detecting the spatial-temporal fluctuations of SMC. The aim of this study was to estimate SMC at different soil depths using very high-resolution unmanned aerial vehicle (UAV)-based multispectral (MS) images and machine learning algorithms and generate spatial maps of SMC using the best-performed machine learning (ML) algorithm. The UAV-based multispectral images of bare soil were captured at 40 m altitude with a very high spatial resolution (2.89 cm) during the rabi 2021/22 season. At the same time, the soil samples were collected from different soil depths, and the gravimetric SMC was measured. Five machine-learning algorithms (Linear Regression (LR), K-Nearest Neighbors (KNN), Random Forest (RF), Decision Tree (DT), and Support Vector Regression (SVR)) were used to train the model between SMC and MS data (MS band reflectance and vegetation indices). The soil with high SMC has low spectral reflectance and soil with low SMC shows high spectral reflectance. For the prediction of surface SMC, the linear regression (R2 = 0.89; RMSE = 2.80 %) and 5 cm depth SMC, the SVR (R2 = 0.64; RMSE = 3.03 %) were performed well compared to other ML algorithms. For surface SMC, blue band reflectance, and 5 cm depth SMC, the Ratio Vegetation Index (RVI) correlated well compared to others. All models failed to predict the SMC at the deeper soil depths. The spatial SMC mapping described the visual color variations in SMC within the field. Crop irrigation scheduling can be significantly improved through the insights this spatial SMC estimation approach provides, making it a valuable tool for farmers and irrigation planners.

Globally quantitative analysis of the impact of atmosphere and spectral response function on 2-band enhanced vegetation index (EVI2) over Sentinel-2 and Landsat-8

The 2-band enhanced vegetation index (EVI2) was developed as an alternative to the enhanced vegetation index (EVI) for sensors that lack a blue band or for observing bright targets such as thick clouds, snow, and ice. However, research has indicated that EVI2 exhibits inconsistencies with the EVI when applied to non-moderate-resolution imaging spectroradiometer (MODIS) sensors. This study aimed to investigate the sources of these inconsistencies by examining the effects of atmospheric conditions and spectral response functions (differences in band reflectance due to sensor spectral response functions) on Landsat-8 and Sentinel-2 globally and evaluate the stability of EVI2 over snow-covered areas. First, a discrete anisotropic radiative transfer (DART) simulation in a snow-covered area was performed to test the atmospheric resistance and sensitivity of the leaf area index (LAI) to EVI and EVI2. Then, two scenarios were analyzed: an ideal case that investigated the spectral response function effect using simulated band reflectances from the United States Geological Survey (USGS) spectral library and a realistic case that examined the effect of both atmospheric and spectral response function using satellite images from the Google Earth Engine (GEE) globally. Our findings revealed that EVI2 demonstrated greater stability in resisting atmospheric influences and sensitivity to LAI than EVI in snow-covered areas. The blue band exhibited a significantly lower transmittance than the other bands, making it more vulnerable to atmospheric conditions. Furthermore, the discrepancy between the EVI and EVI2 for vegetation may be exacerbated by atmospheric effects, as these tend to overestimate the EVI and underestimate the EVI2. This could be attributed to the net positive atmospheric influence on blue band reflectance. The spectral response function effect analysis indicated that EVI2 values may be lower than EVI for vegetation when using Sentinel-2. This underestimation of EVI2 could be mitigated by modifying the spectral corresponding coefficient (c) to 1.45. Here, we demonstrate the superiority of EVI2 over snow-covered areas and offers reliable EVI2 implementation for Landsat-8 and Sentinel-2 application globally.

Paddy Rice Double-Cropping Field Monitoring via Vegetation Indices with Limited Ground Data—A Case Study for Thapanzeik Dam Irrigation District in Myanmar

This study investigated two popular satellite-derived vegetation indices (VIs), MODIS NDVI and EVI, as tools for monitoring crop growth at the Thapanzeik Dam irrigation district in Myanmar, where quality ground data are difficult to obtain. The time-series analysis for seasonal peak VIs presented a considerable improvement in paddy cultivation during 2001–2020 in the irrigation district during summer and monsoon. Fields outside the district had a much lower improvement ratio. Within the irrigation district, a canal with limited water supply was identified via peak and average VIs evaluation. The evaluation of precipitation impact on crop growth demonstrated an opposite impact on crop growth in summer and monsoon cultivations. Water is one of the limiting factors in summer in the irrigation district; thus, precipitation improves water conditions. However, water is not the limiting factor in monsoon; thus, extra water from precipitation, both hydraulically and meteorologically, negatively impacts crop growth. Compared to NDVI, EVI better captured crop growth in densely vegetated surfaces. Meanwhile, information degradation might have occurred with 250 m EVI, using 500 m blue-band reflectance as an input. Thus, the best vegetation index to use depends on the purpose of monitoring and the site situation.

Damage Analysis of Three Long-Track Tornadoes Using High-Resolution Satellite Imagery

Remote sensing of tornado damage can provide valuable observations for post-event surveys and reconstructions. The tornadoes of 3 March 2019 in the southeastern United States are an ideal opportunity to relate high-resolution satellite imagery of damage with estimated wind speeds from post-event surveys, as well as with the Rankine vortex tornado wind field model. Of the spectral metrics tested, the strongest correlations with survey-estimated wind speeds are found using a Normalized Difference Vegetation Index (NDVI, used as a proxy for vegetation health) difference image and a principal components analysis emphasizing differences in red and blue band reflectance. NDVI-differenced values across the width of the EF-4 Beauregard-Smiths Station, Alabama, tornado path resemble the pattern of maximum ground-relative wind speeds across the width of the Rankine vortex model. Maximum damage sampled using these techniques occurred within 130 m of the tornado vortex center. The findings presented herein establish the utility of widely accessible Sentinel imagery, which is shown to have sufficient spatial resolution to make inferences about the intensity and dynamics of violent tornadoes occurring in vegetated areas.

Mechanical Harvesting Effectively Controls Young Typha spp. Invasion and Unmanned Aerial Vehicle Data Enhances Post-treatment Monitoring.

The ecological impacts of invasive plants increase dramatically with time since invasion. Targeting young populations for treatment is therefore an economically and ecologically effective management approach, especially when linked to post-treatment monitoring to evaluate the efficacy of management. However, collecting detailed field-based post-treatment data is prohibitively expensive, typically resulting in inadequate documentation of the ecological effects of invasive plant management. Alternative approaches, such as remote detection with unmanned aerial vehicles (UAV), provide an opportunity to advance the science and practice of restoration ecology. In this study, we sought to determine the plant community response to different mechanical removal treatments to a dominant invasive wetland macrophyte (Typha spp.) along an age-gradient within a Great Lakes coastal wetland. We assessed the post-treatment responses with both intensive field vegetation and UAV data. Prior to treatment, the oldest Typha stands had the lowest plant diversity, lowest native sedge (Carex spp.) cover, and the greatest Typha cover. Following treatment, plots that were mechanically harvested below the surface of the water differed from unharvested control and above-water harvested plots for several plant community measures, including lower Typha dominance, lower native plant cover, and greater floating and submerged aquatic species cover. Repeated-measures analysis revealed that above-water cutting increased plant diversity and aquatic species cover across all ages, and maintained native Carex spp. cover in the youngest portions of Typha stands. UAV data revealed significant post-treatment differences in normalized difference vegetation index (NDVI) scores, blue band reflectance, and vegetation height, and these remotely collected measures corresponded to field observations. Our findings suggest that both mechanically harvesting the above-water biomass of young Typha stands and harvesting older stands below-water will promote overall native community resilience, and increase the abundance of the floating and submerged aquatic plant guilds, which are the most vulnerable to invasions by large macrophytes. UAV's provided fast and spatially expansive data compared to field monitoring, and effectively measured plant community structural responses to different treatments. Study results suggest pairing UAV flights with targeted field data collection to maximize the quality of post-restoration vegetation monitoring.

Operational Data Fusion Framework for Building Frequent Landsat-Like Imagery

An operational data fusion framework was built to generate dense time-series Landsat-like images by fusing MODIS data products and Landsat imagery. The spatial and temporal adaptive reflectance fusion model (STARFM) was integrated in the framework. Compared with earlier implementations of the STARFM, several improvements have been incorporated in the operational data fusion framework. These include viewing angular correction on the MODIS daily bidirectional reflectance, precise and automated coregistration on MODIS and Landsat paired images, and automatic selection of Landsat and MODIS paired dates. Three tests that use MODIS and Landsat data pairs from the same season of the same year, the same season of two different years, and different seasons from adjacent years were performed over a Landsat scene in northern India using the integrated STARFM operational framework. The results show that the accuracy of the predicted results depends on the data consistency between the MODIS nadir bidirectional-reflectance-distribution-function-adjusted reflectance and Landsat surface reflectance on both the paired dates and the prediction dates. When MODIS and Landsat reflectances were consistent, the maximum difference of the predicted results for all Landsat spectral bands, except the blue band, was about 0.007 (or 5.1% relatively). However, differences were larger (0.026 in absolute and 13.8% in relative, except the blue band) when two data sources were inconsistent. In an extreme case, the difference for blue-band reflectance was as large as 0.029 (or 39.1% relatively). Case studies focused on monitoring vegetation condition in central India and the Hindu Kush Himalayan region. In general, spatial and temporal landscape variation could be identified with a high level of detail from the fused data. Vegetation index trajectories derived from the fused products could be associated with specific land cover types that occur in the study regions. The operational data fusion framework provides a feasible and cost-effective way to build dense time-series images at Landsat spatial resolution for cloudy regions.

Improved Accuracy of Chlorophyll-a Concentration Estimates from MODIS Imagery Using a Two-Band Ratio Algorithm and Geostatistics: As Applied to the Monitoring of Eutrophication Processes over Tien Yen Bay (Northern Vietnam)

Sea eutrophication is a natural process of water enrichment caused by increased nutrient loading that severely affects coastal ecosystems by decreasing water quality. The degree of eutrophication can be assessed by chlorophyll-a concentration. This study aims to develop a remote sensing method suitable for estimating chlorophyll-a concentrations in tropical coastal waters with abundant phytoplankton using Moderate Resolution Imaging Spectroradiometer (MODIS)/Terra imagery and to improve the spatial resolution of MODIS/Terra-based estimation from 1 km to 100 m by geostatistics. A model based on the ratio of green and blue band reflectance (rGBr) is proposed considering the bio-optical property of chlorophyll-a. Tien Yen Bay in northern Vietnam, a typical phytoplankton-rich coastal area, was selected as a case study site. The superiority of rGBr over two existing representative models, based on the blue-green band ratio and the red-near infrared band ratio, was demonstrated by a high correlation of the estimated chlorophyll-a concentrations at 40 sites with values measured in situ. Ordinary kriging was then shown to be highly capable of predicting the concentration for regions of the image covered by clouds and, thus, without sea surface data. Resultant space-time maps of concentrations over a year clarified that Tien Yen Bay is characterized by natural eutrophic waters, because the average of chlorophyll-a concentrations exceeded 10 mg/m3 in the summer. The temporal changes of chlorophyll-a concentrations were consistent with average monthly air temperatures and precipitation. Consequently, a combination of rGBr and ordinary kriging can effectively monitor water quality in tropical shallow waters.

Advantages of a two band EVI calculated from solar and photosynthetically active radiation fluxes

A two band Enhanced Vegetation Index (EVI2) without the blue band reflectance has recently been developed as a proxy for the phenology, quantity, and activity of vegetation. We compared the ability of EVI2 and the more commonly used Normalized Difference Vegetation Index (NDVI) to resolve differences in surface greenness and Leaf Area Index (LAI) among three sites located along a burn severity gradient in arctic tundra. We calculated vegetation indices from solar and photosynthetically active radiation fluxes, and validated these calculations against vegetation indices from the Terra MODerate resolution Imaging Spectroradiometer (MODIS) and ground-based spectroradiometer measurements. EVI2 performed slightly better than NDVI when comparing tower derived vegetation indices to MODIS and spectroradiometer derived vegetation indices. Burn severity decreased albedo and resulted in differences in soil background reflectance among sites. Soil darkening had no effect on EVI2, but artificially increased NDVI, resulting in separate relationships between NDVI and Leaf Area Index for burned and unburned tundra. Our results indicate that EVI2 has several advantages over NDVI including the ability to resolve LAI differences for vegetation with different background soil reflectance.

Yellow flowers can decrease NDVI and EVI values: evidence from a field experiment in an alpine meadow

(i) the yellow flower of H. tricuspis significantly increased the red band canopy reflectance, but without apparent variations in near-infrared or blue band reflectance; (ii) the increase in red band reflectance significantly reduced the NDVI and EVI values; and (iii) the EVI was more sensitive than the NDVI to changes in the flower cover. Furthermore, clear decreases were observed in the NDVI and EVI values of the Moderate Resolution Imaging Spectroradiometer (MODIS) product for the study site caused by flower effects. These results suggest that there is a need to examine the effects of flower cover on the NDVI and EVI when they are used for types of vegetation that have profuse seasonal blooms.

Evaluating MODIS, MERIS, and VEGETATION vegetation indices using in situ measurements in a semiarid environment

New and improved satellite sensors specially designed for vegetation monitoring have been launched in recent years; including the Moderate Resolution Imaging Spectrometer (MODIS) onboard Terra and Aqua, the Medium Resolution Imaging Spectrometer (MERIS) on the ENVISAT satellite, and VEGETATION onboard the Systeme Pour l'Observation de la Terre (SPOT) satellite. The aim of this paper is to evaluate two different vegetation indices of these new sensors; the normalized difference vegetation index (NDVI) and the enhanced vegetation index (EVI). This is done by radiometric in situ measurements covering natural grass savanna in Senegal. Variations in the dynamic range of in situ NDVI was found caused by the differences in spectral response functions, MERIS NDVI characterized by the largest dynamic range. Both daily MERIS and MODIS NDVI mirrored accurately in situ measured NDVI (MERIS r2 =0.97 and MODIS r2=0.96). VEGETATION NDVI (only available as ten-day composites) was found to be significantly lower than MODIS NDVI due to lower VEGETATION near-infrared (NIR) reflectance values. A good agreement between the NDVI/EVI relation from satellite and from in situ measured MODIS NDVI/EVI was found. This indicates an accurate atmospheric correction of the MODIS red, NIR, and blue spectral bands, also confirmed by in situ measured reflectances. EVI is sensitive to variations in blue band reflectance, and the consistency between EVI from the different sensors is reduced when compared to NDVI due to the different atmospheric correction schemes of the blue band. Thus, it is recommended that vegetation index cross-sensor algorithms should be based on NDVI over EVI

IMAGE-PROCESSING SOLUTION TO COTTON COLOR MEASUREMENT PROBLEMS: PART II. INSTRUMENT TEST AND EVALUATION

An experimental cotton color/trash meter was developed previously for the purpose of improving cotton colormeasurement by removing trash-particle effects on color measurement with image processing. In this work, the experimentalmeter was tested extensively on a large number of cotton samples varying widely in color and trash content. Testing involved:(1) comparing the measurement accuracy of the experimental meter to that of conventional meters that use diffuse reflectancefor color measurement, and (2) comparing the experimental and conventional meters ability to predict clean lint color fromthat of uncleaned lint. Results indicated that basic cotton color measurement was as accurate with the experimental meteras it was with conventional meters. Additionally, the experimental meters color measurements on uncleaned lint correlatedbetter with cleaned lint color than did those of the conventional meters in every case. With the Z (blue-band reflectance) colormeasurement, the superiority of the experimental meters correlation between uncleaned and cleaned lint color wasstatistically significant. The reduction in root-mean-square error with the experimental meter was about 14% for the Y(broad-band green reflectance) measurement and about 22% for Z. These reductions in prediction error had statistical as wellas practical significance.

Foreign Matter Effects on Cotton Color Measurement: Determination and Correction

Literature has implied that foreign matter affects cotton reflectance and yellowness measurements made with color meters. Samples of seed cotton and lint were measured for reflectance and yellowness before and after hand cleaning of the sample faces. Results confirmed the existence of the cleaning effect on color measurements which was more pronounced with seed cotton. Because disturbance of the sample face was minimized, change in trash content appeared to be the main cause of the effect. The change in the measured area of trash in the sample face correlated well with the change in reflectance for seed cotton, but change in yellowness was less predictable. To remove the effects of trash particles on color measurements, a new method was devised in which spectrally filtered images of lint and seed cotton samples were acquired with a monochrome video camera, and a computer algorithm was written to allow color measurement on essentially lint only by not considering pixels classified as trash in the calculation of either reflectance or yellowness. Color and trash measurements were also made with a High Volume Instrument color/trash meter. Trash particles then were removed by hand from the faces of the cotton samples. Color and trash measurements were then made again with the High Volume Instrument color/trash meter. While prediction of green-red reflectance after cleaning was not improved with the new method, the blue-band reflectance of uncleaned seed cotton was more highly correlated with that of cleaned seed cotton when the new measurement technique was used. This improvement was also marginally true with lint. Additionally, prediction of blue-band reflectance of cleaned lint from that of uncleaned seed cotton was improved with the new method. Gin process-control systems requiring evaluation of seed cotton color would benefit from the use of such a technique which lessens the effects of nonlint material on color measurements.