drones Article Effect of the Solar Zenith Angles at Different Latitudes on Estimated Crop Vegetation Indices Milton Valencia-Ortiz 1 , Worasit Sangjan 1 , Michael Gomez Selvaraj 2 , Rebecca J. McGee 3 and Sindhuja Sankaran 1, *   Citation: Valencia-Ortiz, M.; Sangjan, W.; Selvaraj, M.G.; McGee, R.J.; Sankaran, S. Effect of the Solar Zenith Angles at Different Latitudes on Estimated Crop Vegetation Indices. Drones 2021, 5, 80. https://doi.org/10.3390/ drones5030080 Academic Editor: Adam T. Cross Received: 29 June 2021 Accepted: 13 August 2021 Published: 18 August 2021 Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affil- iations. Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). 1 Department of Biological System Engineering, Washington State University, Pullman, WA 99164, USA; m.valenciaortiz@wsu.edu (M.V.-O.); worasit.sangjan@wsu.edu (W.S.) 2 Alliance Bioversity International and International Center for Tropical Agriculture (CIAT), Km 17 Recta Cali-Palmira, Palmira 763537, Colombia; m.selvaraj@cgiar.org 3 USDA-ARS, Grain Legume Genetics and Physiology Research Unit, Pullman, WA 99164, USA; rebecca.mcgee@usda.gov * Correspondence: sindhuja.sankaran@wsu.edu; Tel.: +1-509-335-8828 Abstract: Normalization of anisotropic solar reflectance is an essential factor that needs to be consid- ered for field-based phenotyping applications to ensure reliability, consistency, and interpretability of time-series multispectral data acquired using an unmanned aerial vehicle (UAV). Different models have been developed to characterize the bidirectional reflectance distribution function. However, the substantial variation in crop breeding trials, in terms of vegetation structure configuration, creates challenges to such modeling approaches. This study evaluated the variation in standard vegetation indices and its relationship with ground-reference data (measured crop traits such as seed/grain yield) in multiple crop breeding trials as a function of solar zenith angles (SZA). UAV-based multi- spectral images were acquired and utilized to extract vegetation indices at SZA across two different latitudes. The pea and chickpea breeding materials were evaluated in a high latitude (46 ◦ 36 ′ 39.92 ′′ N) zone, whereas the rice lines were assessed in a low latitude (3 ◦ 29 ′ 42.43 ′′ N) zone. In general, several of the vegetation index data were affected by SZA (e.g., normalized difference vegetation index, green normalized difference vegetation index, normalized difference red-edge index, etc.) in both latitudes. Nevertheless, the simple ratio index (SR) showed less variability across SZA in both latitude zones amongst these indices. In addition, it was interesting to note that the correlation between vegetation indices and ground-reference data remained stable across SZA in both latitude zones. In summary, SR was found to have a minimum anisotropic reflectance effect in both zones, and the other vegetation indices can be utilized to evaluate relative differences in crop performances, although the absolute data would be affected by SZA. Keywords: phenomics; unmanned aerial vehicle; multispectral imaging; field crops 1. Introduction Remote sensing applications utilizing light interactions (from visible to shortwave infrared) with crop/plant leaves can be used to monitor crop responses. The biochemical and biophysical factors within a leaf, including cellular organization, determine its optical properties, such as reflectance [1]. However, multi-angular remote sensing can affect the reflectance data due to the anisotropic reflectance property, which affects both the radiance intensity and spectral distribution captured by the sensor [2]. Many approaches to represent anisotropic reflectance using the bidirectional reflectance distribution function (BRDF) have been developed, and it has progressively gained recognition in the remote sensing community [3]. BRDF is crucial for radiometric adjustments, especially in terms of removing disruptions that could limit image interpretation skills [4]. Different mathematical methods based on BRDF have been devised to correct the anisotropic reflectance property [5]. For example, Latifovic et al. [6] compared four BRDF Drones 2021, 5, 80. https://doi.org/10.3390/drones5030080 https://www.mdpi.com/journal/drones