Effect of Time of Day and Sky Conditions on Different Vegetation Indices Calculated from Active and Passive Sensors and Images Taken from UAV

Abstract

Optical sensors have been widely reported to be useful tools to assess biomass, nutrition, and water status in several crops. However, the use of these sensors could be affected by the time of day and sky conditions. This study aimed to evaluate the effect of time of day and sky conditions (sunny versus overcast) on several vegetation indices (VI) calculated from two active sensors (the Crop Circle ACS-470 and Greenseeker RT100), two passive sensors (the hyperspectral bidirectional passive spectrometer and HandySpec Field sensor), and images taken from an unmanned aerial vehicle (UAV). The experimental work was conducted in a wheat crop in south-west Germany, with eight nitrogen (N) application treatments. Optical sensor measurements were made throughout the vegetative growth period on different dates in 2019 at 9:00, 14:00, and 16:00 solar time to evaluate the effect of time of day, and on a sunny and overcast day only at 9:00 h to evaluate the influence of sky conditions on different vegetation indices. For most vegetation indices evaluated, there were significant differences between paired time measurements, regardless of the sensor and day of measurement. The smallest differences between measurement times were found between measurements at 14:00 and 16:00 h, and they were observed for the vehicle-carried and the handheld hyperspectral passive sensor being lower than 2% and 4%, respectively, for the indices NIR/Red edge ratio, Red edge inflection point (REIP), and the water index. Differences were lower than 5% for the vehicle-carried active sensors Crop Circle ACS-470 (indices NIR/Red edge and NIR/Red ratios, and NDVI) and Greenseeker RT100 (index NDVI). The most stable indices over measurement times were the NIR/Red edge ratio, water index, and REIP index, regardless of the sensor used. The most considerable differences between measurement times were found for the simple ratios NIR/Red and NIR/Green. For measurements made on a sunny and overcast day, the most stable were the indices NIR/Red edge ratio, water index, and REIP. In practical terms, these results confirm that passive and active sensors could be used to measure on-farm at any time of day from 9:00 to 16:00 h by choosing optimized indices.

Daytime and seasonal reflectance of maize grown in varying compass directions

“…Our observations agree with a recent report that evaluated daytime changes with different terrestrial and UAV-based sensors and found that spectral indices differed significantly in wheat independent of the sensor platform used (De Souza et al, 2021). However, such differences may be more marked in other row crops because of varying solar angles.…”

“…For mission planning, eMotion 3 and Pix4D software from senseFly (Lausanne, Switzerland) were used to provide an 85% lateral and 95% longitudinal overlap. Further details were reported by De Souza et al (2021). Calibration was performed using a white reference standard to process UAV data for each flight.…”

“…Its potential has rarely been determined across the entire life cycle of plants to determine biomass changes over time and make early yield predictions. Although the first studies on diurnal and seasonal influences on the spectral signature of drones have been reported for wheat (De Souza et al, 2021), such studies are lacking for maize, which is one of the most important crops grown worldwide. This requires highly managed experimental designs to determine the possible daytime and seasonal influences of varying solar radiation and angles on the spectral signatures of maize grown in various compass directions.…”

UAV-Based Estimation of Grain Yield for Plant Breeding: Applied Strategies for Optimizing the Use of Sensors, Vegetation Indices, Growth Stages, and Machine Learning Algorithms

“…Earlier approaches of field phenotyping applied ground-based vehicles [14]. However, recent technological advances and a decrease in costs favor the use of drones, i.e., unmanned aerial vehicles (UAVs) [11,[15][16][17]. Notably, UAVs have the advantage of faster measurement without the risk of damaging trial plots.…”

Utilization of UAV Remote Sensing in Plant-Based Field Experiments: A Case Study of the Evaluation of LAI in a Small-Scale Sweetcorn Experiment

“…The aerial multispectral images were taken using the multispectral imaging sensor "Sequoia" equipped on the UAV "DJI Phantom3 advanced" in 2018 and on the UAV "DJI Mavic Pro" in 2019. Since measurements of passive reflectance sensors are influenced by time of day and solar elevation angle, it requires consideration of the angular variation in reflectance and ambient light fluctuations [31,32]. Therefore, the RGB and multispectral image data collection using UAVs measurements were made close to noon, between 10:00 a.m. and 2:00 p.m. on cloud-free and sunny days.…”