Harnessing Advances in Earth Observation and Machine Learning for Agricultural Informatics
Abstract
A data-driven approach to agricultural production has the potential to deliver a more sustainable utilization and management of resources, while increasing the output of farms from local to global scales. Precision agriculture represents one approach to agricultural optimization, and is based on the concept of providing needed inputs of the right amount, at the right time, in the right place. Remote sensing has an obvious role in advancing this field, but has often been constrained by the temporal frequency and spatial resolution of observations, as well as the latency with which data can be obtained. With advances in CubeSats and UAVs increasingly supplementing more traditional satellite platforms, these spatiotemporal barriers are increasingly being overcome. As a consequence, we are awash with rich geospatial datasets. However, turning this information into useable and useful products remains an ongoing challenge, particularly for many emerging sensor technologies, whose instruments may not be as rigorously calibrated as space agency systems. Machine learning approaches, together with the rise of cloud computing and the data-cube, have enabled a dramatic shift in how such information can be obtained, interrogated and disseminated. These new resources provide opportunities to advance not just precision agriculture, but also offer tools and information that is needed to address food and water security challenges. This talk will present an overview of some recent efforts to develop useable products from a range of earth observing platforms, and the application of novel machine learning approaches that are enabling the production of scalable, field-level, agricultural informatics.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2018
- Bibcode:
- 2018AGUFM.H33A..02M
- Keywords:
-
- 0434 Data sets;
- BIOGEOSCIENCESDE: 1855 Remote sensing;
- HYDROLOGYDE: 1926 Geospatial;
- INFORMATICSDE: 1942 Machine learning;
- INFORMATICS