Figure 13: Cumulative heat storage in the upper 70 m of the water column for years 2018, 2020 and 2021. There are no data shown for 2019 due to a major thermistor chain malfunction. Calculations start from 24 June as no data were recorded before this date for 2018 and 2020. For the purposes of the figure, to create a reference point based on our data availability, we set the annual cumulative heat storage at the end of the year to zero (ΔHS = 0 MJ m−2).
Note that Romaine-2 experienced a sharp drop in heat storage in September 2020. This is linked to the approximately 10 m decline in water level in August and September. Indeed, the epilimnion moved closer to the water intake and therefore the accumulated heat within the top layers was drained out as turbine flow. The year 2020 was therefore unusual in terms of heat storage.
When looking at the monthly layer energy balance of the reservoir using equation 3, we noted that there was a significant non-closure. This is due to the time scale difference between the heat storage in the water column and the remaining terms (turbulent heat fluxes and net radiation). Indeed, due to the high thermal inertia and the high specific heat of water, the net radiation that was received in summer accumulated in the water column and was only released in autumn. This feature makes it difficult to calculate the EBR on a monthly scale. Within this context, correcting turbulent heat fluxes to account for the energy imbalance (Foken, 2008) using the monthly EBR is problematic, unless accurate measurements of lateral energy inputs are available (sect. 3.3).