Assessment of the spatiotemporal variability of trees status in an olive orchard through multispectral drone images

• Published in: African and Mediterranean Agricultural Journal - Al Awamia no. 132; pp. 142 - 162
• Main Authors: S. El Iraqui El Houssaini, H. Iaaich, A. El Bakkali, R. Razouk, K. Bouhafa, K. Habbadi, A. Benbouazza, M. Alghoum, A. Douaik
• Format: Journal Article
• Language: English
• Published: National Institute of Agronomic Research "INRA" Morocco 01.03.2022
• Subjects: agriculture de précision; gndvi; indices de végétation; maroc; ndre; ndvi; olivier
• ISSN: 2658-9184
• DOI: 10.34874/IMIST.PRSM/afrimed-i132.31336

Precision Agriculture (PA) is a management strategy based on Information and Communication Technologies (ICT) used to assist management decisions with the aim to increase productivity, to reduce labor time and inputs needed, to improve benefits and to protect the environment. Techniques related to PA were initially used to monitor status of annual crops and they were recently adopted on fruit trees. Nevertheless, PA is still in its initiation phase in Morocco. Given the importance of olive in the national and international levels, applying this concept in the intensive and hyper-intensive production systems would be a reliable and cost-effective approach for the assessment of orchards health and nutritional status and then make reliable decisions. With the aim to assess the potential of multi-spectral images to inform on temporal and spatial variability characterizing an olive orchard, this study was carried out in an orchard of 22 ha, in Meknes region, Morocco, planted principally with “Picholine marocaine” variety. Four Unmanned Aerial Vehicle (UAV) flights, from June to November 2020, were performed and three vegetation indices (VIs) were computed: NDVI, GNDVI and NDRE. Spatial variability was assessed using variogram, a geostatistical tool, and temporal variability was evaluated using Spearman rank correlation and relative difference methods. The three VIs showed a weak temporal and spatial stability of trees which indicate an overall unstable canopy condition related to tree physiological behavior and environmental conditions (temperature and rainfall). Additionally, the observation of a potential stress was recorded firstly by NDRE, followed by NDVI and GNDVI. Results show promising perspectives of using remote sensing techniques to assess olive trees status and therefore support farmers in orchards management. Such challenge requires more in-depth work to develop in house models correlating VIs to agronomical parameters.