Original paper

Empirical predictions of water temperatures in a subtropical reservoir

Groeger, Alan W.; Bass, David A.


A series of empirically-derived equations were developed from data spanning the 32 year history (1969 – 2000) that predict or reconstruct water temperatures of the inflowing river, the near surface or epilimnion, and the deep and outflowing waters of subtropical Canyon Reservoir on the Guadalupe River in southcentral Texas, USA. Inflowing and near surface water temperatures both can be predicted from local air temperatures. We used a 5 day or 1 day (in the case of high inflows associated with storm fronts) averaging period of air temperature for the river and a 30 day period for the reservoir surface water. Predictions of deep water and outflow temperatures were more complex in this bottom-draining reservoir. During the winter months, when the reservoir was nearest to being isothermal, air temperature was the primary factor used to predict deep temperatures. For the remainder of the year, the rate at which water was released from the reservoir tended to be the most important factor in predicting deep temperatures, although the air temperature of the previous winter could also be important. During drier years, the cold hypolimnetic waters are retained throughout the stratified period, and outflows were as cool as 12 –13 °C in August and September. During very wet years the cold hypolimnetic waters are lost downstream early, and August and September outflows may be as warm as 23 – 24 °C. Therefore, interannual variation in the temperature of the hypolimnion and tailwaters is quite high. This represents a thermal stress to the local biota that is very different from the idea of "thermal constancy" commonly associated with other reservoir tailwater habitats. Because of this reservoir's particular sensitivity to variation in flow, projections of future change in runoff (as the climate changes) may have a greater effect on the thermal environment of this reservoir than will change in atmospheric temperature.


predictive modellong-termtemperature-variationclimate change