Automated site-specific irrigation framework and evaluation
Presentation
Paper/Presentation Title | Automated site-specific irrigation framework and evaluation |
---|---|
Presentation Type | Presentation |
Authors | |
Author | McCarthy, Alison |
Journal or Proceedings Title | Proceedings of the Irrigation Australia International Conference and Exhibition 2016 |
Year | 2016 |
Conference/Event | Irrigation Australia International Conference and Exhibition 2016: Irrigation - For Prosperity and Wellbeing |
Event Details | Irrigation Australia International Conference and Exhibition 2016: Irrigation - For Prosperity and Wellbeing Event Date 24 to end of 26 May 2016 Event Location Melbourne, Australia |
Abstract | The spatial variability of plant available water content and irrigation requirement within a single field can be up to 200%. Measuring this variability typically requires either aerial vehicles (remotely piloted or piloted) with cameras and optical and depth sensors, or ground vehicles (e.g. tractors, mopeds) carrying electrical conductivity and optical sensors over the field. However, conducting these surveys multiple times during the season is time consuming. Alternatively, a fixed sensor could be installed in each management zone in the field, and the infield variability estimated using spatial interpolation. Fieldwork and spatial analysis has been completed to identify the errors in the spatial interpolation and impact on irrigation requirement calculated. Fieldwork was conducted on 5 ha of two surface-irrigated cotton fields in 2014/15 in Yargullen, Queensland. A weather station and fixed soil moisture sensors were installed. Fortnightly spatial measurements of soil electrical conductivity, Normalised Difference Vegetation Index, Ratio Vegetation Index, optical reflectance and plant parameters extracted from camera images were collected. A spatial analysis was conducted to compare measured spatial variability and variability estimated using spatial interpolation between 1, 4, 10 and 50 locations in the field. The interpolation error reduced as the number of data points in the field increased. An artificial intelligence-based analysis was also conducted to identify the sensor locations that minimised error in interpolation for both soil and plant variability. This analysis indicated that there was only a 30% overlap in optimal locations for both soil and plant parameters. |
Keywords | variable-rate; automation; machine vision |
ANZSRC Field of Research 2020 | 300206. Agricultural spatial analysis and modelling |
400799. Control engineering, mechatronics and robotics not elsewhere classified | |
300403. Agronomy | |
Byline Affiliations | Institute for Agriculture and the Environment |
Institution of Origin | University of Southern Queensland |
https://research.usq.edu.au/item/q375y/automated-site-specific-irrigation-framework-and-evaluation
1820
total views44
total downloads0
views this month0
downloads this month