Informations et ressources scientifiques
sur le développement des zones arides et semi-arides

Accueil du site → Doctorat → Australie → 2006 → Use of remote sensing data for broad acre grain crop monitoring in Southeast Australia

Monash University (2006)

Use of remote sensing data for broad acre grain crop monitoring in Southeast Australia

Coppa, I

Titre : Use of remote sensing data for broad acre grain crop monitoring in Southeast Australia

Auteur : Coppa, I

Université de soutenance : Monash University

Grade : Doctor of Philosophy (PhD) 2006

In 2025, there will be almost 8 billion people to feed as the worlds population rapidly increases. To meet domestic and export demands, Australian grain productivity needs to approximately triple in the next 20 years, and this production needs to occur in an environmentally sustainable manner. The advent of Hi-tech Precision Farming in Australia has shown promise in recent time to optimize the use of resources. Most "precision farmers" produce yield maps at harvest. However when yield maps become available it is usually too late to apply management techniques that would address problems in the crop specific to the current season. A satellite-based broad-acre crop monitoring system could provide intelligence on crop performance throughout the growing season. A prototype concept for a system that used commercially available satellite imagery was developed in this research, named ALMIS (Agricultural Land Management Information System) and tested with 25 farmers in the Gooroc area in Victoria with a focus on the following crop types ; barley, canola, chickpeas, lentils and wheat. Through the study, several components providing vital information for a crop monitoring system were developed from satellite imagery, including crop type discrimination, quantitative crop parameters, crop yield estimates. Furthermore, critical parameters for the system were determined.
The feedback given by the Gooroc farmers, together with the results gained from the ALMIS field studies and considerations reported in the literature were the foundations for the concept design proposal. As critical parameters were identified (amongst others) : delivery of data within 1-2 days after acquisition ; data should have spatial resolution better than 10 meters, and appropriate geometric and radiometric stability ; in an easy-to-use computer interface (software or via internet), carefully timed image data should be offered as near infrared imagery, colour-coded vegetation index, percentile rating for each paddock, quantitative vegetation parameter and yield forecast maps ; assistance should to be given to farmers by trained agronomists to convert information into management decisions, and training courses are needed to give end users basic understanding of the technologies utilized by such a system ; consolidated strategic development efforts need to be employed to further the use of the information in decision support systems and variable rate technology. As the technology is in its infancy it needs to be supported by government and industry initiatives to reach critical mass. _It was concluded that information from satellite remote sensing is greatly beneficial to the farming community of south east Australia, resulting in substantial economic benefits for local farmers as it becomes widely adapted. The technology developed in this thesis contributes to the Australian goal of increasing crop yields in a profitable and environmentally friendly manner.


Version intégrale (5,58 Mb)

Page publiée le 7 novembre 2019