OnePage Bootstrap Template - Index

Identifying drought predictors that best predict regional water shortages

Drought conditions produce large economic impacts in water-limited communities, and are becoming more frequent and severe. When productivity of agricultural, municipal, and industrial operations are limited by available water, drought is a significant risk. One way to understand drought’s impacts on these communities is quantifying administrative water shortage – or the difference between water demand and allocated supply – and its sensitivity to climatic variables that are used to describe drought.

Lynker scientists evaluated water districts in the South Platte and Colorado River Basins in the State of Colorado to explore the sensitivity of water shortages to drought conditions. We created this interactive dashboard for you to explore yourself.

Univariate Linear Regression

How do climate variables relate to shortages?

Multiple Linear Regression

How sensitive are districts to climate? What combination of drought predictors best predict water shortages?

Water Rights

How do water shortages at individual diversion structures change over time?

Climate Change Impact

How will climate change affect water shortages in the future?




How it works:

This dashboard uses supply and demand data from water rights using Statemod, a water allocation model developed by the State of Colorado to represent water rights operations within the state’s unique and carefully managed administrative regime. Statemod has more than 500 model nodes in both the Colorado and South Platte river basins.

Climate data is from the gridMET climate dataset which contains daily gridded climate data at a ~4km spatial resolution for the period 1980 – 2012. These data are used in various ways to measure or understand drought and drought severity. Types of data include hydroclimate data and drought indexes.

Data is aggregated at a district level. 

Hydroclimate data and Drought Indexes

  • Temperature  (C)
  • Precipitation (mm)
  • Snow water equivalent (SWE)
  • Potential Evapotranspiration (mm)​
  • Soil Moisture (mm)
  • Palmer Drought Severity Index
  • Evaporative Demand Drought Index (1, 3, 6, 9, and 12 month)
  • Standard Precipitation Index (1, 3, 6, 9, and 12 month)

Univariate Linear Regression

We can quantify the strength of the relationship between a single climate predictor and water shortages (the explanatory variable) by performing a linear regression analysis. ​ For example, it is intuitive that if precipitation in a district decreases, then the water available for use drops, creating a mismatch between supply and demand. The slope of this line (the strength of the relationship) varies by district.​ Explore all the other relationships between climate predictors and water shortages here.

Multiple Linear Regression

Using a multiple linear regression (MLR) approach, we found the climate indicators that best predict water shortages (demand – supply) for each district. District by district, you can see the climate predictors that have been included in the model

Water Rights

Water districts have a unique portfolio of water rights, consisting of irrigators, municipalities, and other users. This tab explores the diversion locations, demands, and direct flow shortages (demand – direct flow supply) at each headgate.​
Water rights with later more recent appropriation dates are more likely to experience water shortages than senior water rights with early appropriation dates.

Climate Change Impact

​ Using climate models centered around the year 2050, we can create new climate data inputs. Then, using our multiple linear regression models (see first tab), we can see how shortages can be impacted by a shift in climate data.​