Application of a Runoff Temperature Model (MINUHET) to a Residential Development in Plymouth, MN

Abstract

The MINUHET (MINnesota Urban Heat Export Tool) model is a simulation tool used to route heat and storm water through a sub-watershed for a rainfall event or events of interest. The model includes components for developed land uses, undeveloped or vegetated land uses, pervious and impervious open channels, storm sewer systems, and storm water ponds. As a case study, the model has been applied to a 12.5 acre housing development in Plymouth, MN. The process of identifying necessary data is outlined, as well as a general strategy for organizing the input data and setting up the model for this particular watershed. A catch basin at the outlet of the development was instrumented for flow and temperature, and data were collected at the site from August 25, 2005 to October 1, 2005. The model was run for three rainfall events, and a comparison was made between observed and simulated flow rate and flow temperature at the development outlet. Overall, the model performed well. The RMSE for flow was 42.0 L/s, 10.4 L/s, and 14.3 L/s for the three events respectively, and the corresponding RMSE in storm water runoff temperature was 1.6 °C, 1.2 °C, and 1.9 °C. Observed and simulated volumeaveraged mean runoff temperature differed by less than 1.5 ºC for all three events. Total volume of runoff was predicted with reasonable accuracy by the model, especially for the first two events. Heat export, which is a measure of the heat content of the runoff above a certain reference temperature (in this case 16.0 °C), was accurately predicted for the second and third events. The model was found to be highly sensitive to saturated hydraulic conductivity and rainfall temperature (dew point temperature): volume of runoff from the pervious areas varied considerably with changes in hydraulic conductivity, and runoff temperature often tended toward dew point temperature, especially in the absence of large atmospheric or ground heat fluxes (e.g., late at night or early in the morning). This suggests that special care should be taken in selection of soil properties, and that all climate data should be collected as near to the study site as possible to improve the accuracy of runoff temperature estimation.