Farming systems monitoring using machine learning and trend analysis methods based on fitted NDVI time series data in a semi-arid region of Morocco

Y. Lebrini; T. Benabdelouahab; A. Boudhar; A. Htitiou; R. Hadria; H. Lionboui

doi:10.1117/12.2532928

21 October 2019 Farming systems monitoring using machine learning and trend analysis methods based on fitted NDVI time series data in a semi-arid region of Morocco

Y. Lebrini, T. Benabdelouahab, A. Boudhar, A. Htitiou, R. Hadria, H. Lionboui

Author Affiliations +

Proceedings Volume 11149, Remote Sensing for Agriculture, Ecosystems, and Hydrology XXI; 111490S (2019) https://doi.org/10.1117/12.2532928
Event: SPIE Remote Sensing, 2019, Strasbourg, France

Abstract

Managers and policy makers demand information on agriculture dynamics and distribution for the establishment of plans and strategies. For this purpose, the use of remote sensing data constitute an essential key to follow-up the agricultural systems dynamics. The aim of this study is to define a method based on fitted Normalized Difference Vegetation Index (NDVI) time series extracted from Moderate Resolution Imaging Spectroradiometer (MODIS), trend analysis tests and machine learning approaches for assessing and monitoring farming systems in a semi-arid region of Morocco. NDVI time series were smoothed using TIMESAT software for the period between 2000 and 2018. Then, three trend analysis tests were conducted which are: monotonic trend (Mann-Kendall), Man-Kendall significance and median trend (Theil-Sen). In addition, Random Forest (RF) classification methods were performed to classify the main agricultural cover type over the study area for the 2017/2018 cropping season. The results demonstrated the ability of fitted NDVI data and RF classification to identify the main agricultural systems, which are: 1) irrigated annual crop, 2) irrigated perennial crop, 3) rainfed areas and 4) fallow. Analysis of trend patterns based on fitted NDVI values shows high variability over the farming systems. Irrigated annual and perennial crops present high improvement of biomass activity with a small inter-variability with significant trend. For the Rainfed area and fallow, these classes show a non-significant trend with low degradation of productivity. In addition, these results can constitute a relevant means of control and spatio-temporal monitoring of farming systems. Overall, the results are relevant for managers and policy makers to develop procedures and actions in order to prevent environmental and agricultural events resulted from the spatio-temporal changes in farming systems.

Citation Download Citation

Y. Lebrini, T. Benabdelouahab, A. Boudhar, A. Htitiou, R. Hadria, and H. Lionboui "Farming systems monitoring using machine learning and trend analysis methods based on fitted NDVI time series data in a semi-arid region of Morocco", Proc. SPIE 11149, Remote Sensing for Agriculture, Ecosystems, and Hydrology XXI, 111490S (21 October 2019); https://doi.org/10.1117/12.2532928

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available