Micro-Watershed Forecasting Under Uncertainty

Project Description

The majority of water used in the state of Colorado is diverted and delivered by mutual ditch and irrigation companies. Holders of private water rights, irrigation districts, and conservancy districts, are responsible for delivering additional water to farms, industry and municipalities.  The mutual ditch and irrigation companies (ditch companies) are the key social organizational link between the source of water for irrigation and its application on the farm.  There are estimated to be well over 200 major water supply organizations of this nature in Colorado alone, with perhaps an additional 300+ smaller entities affiliated with these 200 major organizations.  What these organizations know, use and adopt in the way of information technology is central to improving water conservation, production, and financial viability in irrigated agriculture.  These water supply organizations are often the best starting point for technology transfer in irrigated agriculture. 

It is argued that future research needs to provide more capabilities for irrigators in site-specific forecasting, rather than in large river basin or regional forecasting. Climate forecasting information designed for use by agricultural water supply organizations has lagged behind other information and hardware technologies for the distribution and management of water, such as technologies for improved business record keeping and improved measurement of water to achieve equity in water distribution and water conservation.  The use of decision strategies surrounding information on climate factors may be the most important and underutilized tool that these water supply organizations have at their disposal today. Their forecasts may have a profound influence on the financial viability of their farmer shareholders.  If improvements can be made in determining the probability of a quantity of water from their organizations, irrigators can then explore an optimal portfolio of crops, reduce risk in the extent of plantings, determine more appropriate irrigation scheduling, and participate more effectively in emerging water marketing institutions (e.g., forbearance contracts, water banking, informal exchanges, leasing of water, etc.). 

The research will take place in the context of working with, learning from, and partnering with local ditch companies and the Ditch and Reservoir Company Alliance of Colorado (DARCA).  This “ownership” in the research by the ditch companies and a reputable representative of their interests (DARCA) is crucial for the successful adoption of the output of the research.

The research will develop a Monte Carlo simulation model that allows mutual ditch and irrigation companies and irrigation districts to better estimate the quota of water they announce at the beginning of the irrigation season for irrigators served by them.  Except for the smallest private ditches in irrigated agriculture, a quota is normally announced by these water supply organizations indicating the amount of water per share of stock (for ditch companies), or per acre of land (for irrigation districts) that irrigators can expect to receive during the irrigation season.  The quota is usually announced in acre-feet or inches of water to be received by the irrigator.  The quota is usually announced in conjunction with the date of first water delivery by the organization as well.

The analysis and output will be performed in the context of risk. The inputs into the model and its results will be described non-parametrically.  The methodology includes coupling regression techniques with Monte Carlo simulation and using forecasts that involve the reporting of statistical results in terms of likelihoods.  The available amount of water will be forecasted - not as a single point estimate but rather as a distribution. Statistics will be derived from distributions including percentiles that can be used more beneficially than single numbers that lack confidence intervals.  In essence, the likelihood of available water will be predicted and then the growers within the water supply organization can make their decisions based on how much risk they are willing to undertake.