2025 Orange County Water Demand Projection Model
6.1.2
Multiple Dry Year Scenario
Trends in historical regional water use (supplied from all sources) were correlated with trends in observed weather conditions to develop a set of factors to describe the potential impact of consecutive dry years. The weather coefficients of the sectoral water demand equations can be used to generate weather scenarios for any given set of monthly weather data, for example, for a given hot/dry year. However, conceptually, the persistence of drier-than-normal weather could intensify water use rates in the absence of interventions, such as water-use restrictions. Because the sectoral forecasting equations, by construction, treat time periods independently, other statistical methods were derived to evaluate potential or “latent”, or unconstrained, demands that could develop with the persistence of dry weather conditions. Trends in historical total regional water use (supplied from all sources) were correlated with trends in observed Orange County weather conditions to develop a set of factors to describe the potential impact of consecutive dry years. Specifically, the running 12-month average of regional water use (USE12) was modeled as a function of the following variables: 12-month running average of the ratio of observed to normal average maximum daily temperature (MAXT12) 24-month running average of the ratio of observed to normal precipitation (PRCP24) 36-month running average of the ratio of observed to normal precipitation (PCRP36) 48-month running average of the ratio of observed to normal precipitation (PCRP48) 60-month running average of the ratio of observed to normal precipitation (PCRP60) Departure from long-term GDP trend Linear time trend counter (TREND) Except for the linear time counter, all variables were transformed into natural log form prior to estimating the model using ordinary least squares regression. A binary water restriction variable was added to the regression to eliminate supply constraints (allow the regression to assume no curtailments instated during the hot dry spell). The estimated weather parameters ( Table 6-2 ) are used to estimate the potential change in water use that would occur during the driest 24-month, 36-month, 48-month, and 60-month period over the historical record, assuming the warmest 12-month period estimated over the historical record.
Table 6-2: Regression Coefficients and Variable Values
Coefficient
Historical Value
MAXT12 PRCP24 PRCP36 PRCP48 PRCP60
0.6279
1.0522 0.3783 0.4408 0.4977 0.5656
-0.0832 -0.0076 -0.0269 -0.0296
6-3
Appendix G - 72
Powered by FlippingBook