Agritrop
Accueil

Forecasts of desert locust presence in Morocco coupling remote sensing imagery and field surveys

Smiej Mohammed F., Layelmam Mohammed, Atillah Abderrahman, Piou Cyril, Ghaout Saïd. 2019. Forecasts of desert locust presence in Morocco coupling remote sensing imagery and field surveys. In : 13th International Congress of Orthopterology: Abstract book 2019. University Ibn Zohr, National Center for Control of Desert Locust, Orthopterists Society. Agadir : University Ibn Zohr, Résumé, p. 36. (Metaleptea) International Congress of Orthopterology. 13, Agadir, Maroc, 24 Mars 2019/28 Mars 2019.

Communication avec actes
[img]
Prévisualisation
Version publiée - Anglais
Utilisation soumise à autorisation de l'auteur ou du Cirad.
PUB690.pdf

Télécharger (1MB) | Prévisualisation

Résumé : With the objective of improving preventive management of desert locust, an operational system was developed to help in the planning of field surveys in Morocco. This operational system produce regularly some presence probability maps of solitarious or transiens desert locust. The spatial resolution is 25km over the Moroccan territory and the temporal horizon of the forecasts are 40 days. The forecasts are based on statistical models coupling historical data of field surveys with several layers of remote sensing imagery. These images are proxy of environmental variables important for desert locust: temperature, rainfall and vegetation availability. The statistical coupling was realised with random forest models. These models were assessed with a splitting of the data to evaluate the forecast errors and validate the approach. An automatic process was also developed to transform new remote sensing imagery into probability maps in order to operationalize the system. As the system has been running for over 3 years, another level of evaluation can be presented: the correspondence between the forecasts of probability of locust presence and the actual observations of field survey teams of the national anti-locust centre of Morocco since 2015.

Mots-clés libres : Forecasting tool, Random forest models, Remote sensing, Schistocerca gregaria

Auteurs et affiliations

  • Smiej Mohammed F., CTRS (MAR)
  • Layelmam Mohammed, IAV Hassan II (MAR)
  • Atillah Abderrahman, CTRS (MAR)
  • Piou Cyril, CIRAD-BIOS-UMR CBGP (FRA) ORCID: 0000-0002-9378-9404
  • Ghaout Saïd, CNLA (MAR)

Source : Cirad-Agritrop (https://agritrop.cirad.fr/591928/)

Voir la notice (accès réservé à la Dist) Voir la notice (accès réservé à la Dist)

[ Page générée et mise en cache le 2019-10-08 ]