3.2.3 Annual Scale
Calculating annual reservoir evaporation values is challenging. It is well understood that eddy covariance data underestimate turbulent heat fluxes (Foken, 2008), and in addition, the calculation of the heat storage term is especially difficult for a deep water column. For this reason, an alternative approach was used: the corrected turbulent heat fluxes were obtained by redistributing the missing energy according to the annual EBR over a so-called ”energy year,” from March 1 to February 28 of the next year. Since energy storage is typically at its minimum value at this time of year, this allowed us to discard that variable and make Rn = H +LE . SinceRn > H + LE , the missing energy can then be redistributed by preserving the observed Bowen ratio, as discussed in Mauder et al. (2018). By doing so, the EBR was 80%, 69%, and 76% for the years 2019−20, 2020−21, and 2021−22, respectively.
Figure 14 illustrates the yearly cumulative evaporation for three energy years, from 2019 to 2022. The mean annual non-corrected evaporation was 439 ± 23 mm and did not vary much from year to year. When correcting for the energy imbalance using the annual EBR values, the annual evaporation values reached 555 mm, 656 mm and 559 mm for 2019−20, 2020−21 and 2021−22, respectively (refer to Table 1). Note that the inter-annual variabilities of cumulative evaporation were 7.2% and 18% for non-corrected and corrected values, respectively. In fact, 50% of the total measured evaporation occurred over 24%, 27% and 26% of the days for 2019−20, 2020−21 and 2021−22, respectively. This is consistent with Blanken et al. (2000), who found a mean value of 22.5% for the Great Slave Lake.