A new empirical procedure for estimating intra-annual heat storage changes in lakes and reservoirs: Review and analysis of 22 lakes

- A dataset of heat storage changes in 22 lakes was created from literature.
- Heat storage changes can be estimated from net all wave radiation.
- The new hysteresis model has been tested with data from 22 lakes.
- The proposed procedure only requires net radiation and water surface temperature.
- Satellite measurements can provide all required input parameters.

Evaporation is an important component of the water and energy balance of lakes and reservoirs. The energy balance combination method for estimating evaporation requires the heat storage changes (Qt) to be known. The lack of data on water temperature profiles and water depth fluctuations hinders routine computations of Qt. Following successful estimation of the soil heat flux density of land surfaces (G) and heat storage change of urban areas and wetlands from net all wave radiation (Rn), we investigated in this paper whether a similar generic Qt(Rn) empirical relationship can be developed for lakes and reservoirs. A comprehensive literature survey was conducted and experimental datasets for 22 lakes with different characteristics were collected. A linear Qt(Rn) model with a hysteresis function to describe seasonal warming and cooling effects was developed (Qt = a ∗ Rn + b + c ∗ dRn/dt) that fits the 22 independently gathered datasets satisfactory (R2 of 0.83 and RMSE of 22 W m− 2) for bi-weekly and monthly time scales. Predictive models for the coefficients a, b and c were also developed, using Rn and water surface temperature measurements that can be retrieved from routine earth observation measurements. The average R2 between measured and modeled Qt was 0.84 and the RMSE was 37 W m− 2 if predictive models were used for the assessment of lake specific Qt(Rn) functions. Two independent satellite-derived products were explored: the ARC-Lake (ATSR Reprocessing for Climate) product for water surface temperature using the ATSR (Along Track Scanning Radiometer) series data, and the CM SAF (Satellite Application Facility on Climate Monitoring) product for solar radiation Rs based on the AVHRR (Advanced Very High Resolution Radiometer) data. The proposed procedure using purely satellite-derived data as inputs resulted in comparably good Qt estimates as those using in-situ measurements. The new Qt hysteresis model can thus be applied together with satellite measurements for supporting the computation of evaporation from open water bodies based on energy balance equations.