Related Experiment Video
Updated: Aug 24, 2026

Watershed Planning within a Quantitative Scenario Analysis Framework
Published on: July 24, 2016
An approach for treating the uncertainties in the impact of climate change
Y Gu1, J W Crawford, D Ramanee Peiris
1Climate Change Research Group, Cellular and Environmental Physiology Department, Scottish Crop Research Institute, Dundee, UK, DD2 5DA.
Abstract:
Past accumulated data supported by the predictions of climate models suggest that our world is getting warmer. Scientists are trying to construct mathematical models of both climate and crop systems to identify what types of climate changes could constitute a significant risk or benefit for agriculture. However, due to the many uncertainties regarding these models, it is impossible to make unequivocal predictions. At present, almost all the research in this area is carried out without considering the uncertain nature of the problem. The approach outlined here attempts to find a way to deal with the above uncertainty problem. Artificial intelligence techniques are being developed with the aim of performing inferences based on uncertain information. In our method, causal graphs are used for explicit representation of the relationships between climatic factors and yield. Probabilities are used to express the uncertainties associated with these links, and Bayes' theorem is applied to deal with uncertainty reasonings. This approach has the additional advantage of allowing the prediction to be readily updated as results from improved climate and crop models become available. These opportunities are being evaluated initially by using the model for potato growth developed at the Scottish Crop Research Institute.
Related Concept Videos
What is Climate?
Global Climate Change
Strategies for Assessing and Addressing Confounding
Confounding can be addressed at both the design phase of a study and through analytical methods after data...
Propagation of Uncertainty from Random Error
Microbes and Climate Change
Uncertainty: Overview
