Evaporation from a large lowland reservoir – observed dynamics during a warm summer

Evaporation forms a large loss term in the water balance of inland water bodies. During summer seasons, which are projected to become warmer with more severe and prolonged periods of drought, the combination of high evaporation rates and increasing demand on freshwater resources forms a challenge for water managers. Correct parameterisation of open water evaporation is crucial to include in operational hydrological models to make well supported predictions of the loss of water through evaporation. Here, we aim to study the controls on open water evaporation of a large lowland reservoir in the Netherlands. To this end, we analyse the dynamics of open water evaporation at two locations, i.e. Stavoren and Trintelhaven, at the border of Lake IJssel (1100 km2) where eddy covariance systems were installed during the summer seasons of 2019 and 2020. From these measurements we find that wind speed and the vertical vapour pressure gradient, but not available energy, can explain most of the variability of observed hourly open water evaporation. This is in agreement with Dalton's model which is a well-established model often used in oceanographic studies for calculating open water evaporation. At the daily timescale, we find that wind speed and water temperature are the main drivers in Stavoren. These observed driving variables of open water evaporation are used to develop simple data-driven models for both measurement locations. Validation of these models demonstrates that a simple model using only two variables, performs well both at the hourly timescale (R2 = 0.84 in Stavoren, and R2 = 0.67 in Trintelhaven), and at the daily timescale (R2 = 0.72 in Stavoren, and R2 = 0.51 in Trintelhaven). Using only routinely measured meteorological variables leads to well performing simple data-driven models at hourly (R2 = 0.78 in Stavoren, and R2 = 0.51 in Trintelhaven) and daily (R2 = 0.85 in Stavoren, and R2 = 0.43 in Trintelhaven) timescales. These results for the summer periods show that global radiation is not directly coupled to open water evaporation at the hourly or even daily timescale, but rather wind speed and vertical gradient of vapour pressure are variables that explain most of the variance of open water evaporation. However, when we extend the time series to a complete year, we find a distinct yearly cycle reflecting the yearly dynamics of global radiation. We find that the commonly used model of Penman (1948) produces results that resemble the yearly cycle of observed evaporation. However, at the diurnal scale estimated evaporation using Penman’s model disagrees with observed evaporation. Therefore, using the Penman equation to model open water evaporation for shorter periods of time is questioned. We would like to stress the importance of including the correct drivers in the parameterization of open water evaporation in hydrological models to adequately represent the role of evaporation in the surface-atmosphere interaction of inland water bodies.

Evaporation from a large lowland reservoir – observed dynamics during a warm summer

<p>Distinct differences in surface characteristics between a water body and a land surface result in different drivers of evaporation and therefore its dynamics. It is essential to include and represent this difference in the parameterization of open water evaporation (E<sub>water</sub>) to improve operational hydrological models. Additionally, more accurate parameterization becomes even more crucial to predict potential changes in quantity and dynamics of E<sub>water</sub> in a changing climate in support of optimal water management now and in the future.</p><p>For this purpose, we performed a long-term measurement campaign to measure E<sub>water</sub> and related meteorological variables over a large lowland reservoir in the Netherlands. During the summer seasons of 2019 and 2020 eddy-covariance systems were applied at two locations at the border of lake IJsselmeer in the Netherlands. These high temporal resolution measurements gave us the opportunity to explore the dynamics and identify the underlying driving mechanisms of E<sub>water</sub>. Using the data collected during the summer of 2019 we were able to develop a simple regression model for both measurement locations. Combinations, both sums and products, of the following independent variables were considered: global radiation, wind speed, water skin temperature, vapour pressure deficit, and vertical vapour pressure gradient. The product of wind speed and vertical vapour pressure gradient best explained the observed hourly E<sub>water</sub> rates, which is consistent with the commonly used aerodynamic approach. The model was validated using the data of 2020. Additionally, we compared measured E<sub>water</sub> to E<sub>water</sub> computed with Makkink’s equation, which is currently used in the Dutch operational hydrological models to estimate E<sub>water</sub>. Although a correction factor is applied to account for the difference between land evaporation and E<sub>water</sub>, Makkink is not able to capture the dynamics of E<sub>water</sub>. This was reflected in the timing and shape of the evaporation peak at both daily and monthly scales. The disagreement of E<sub>water</sub> dynamics found between the measured and simulated E<sub>water</sub> even more demonstrates the value and need of a correct parameterization of E<sub>water</sub>.</p>

Floristic diversity in the small riverswith different morphology in the zoneaffected by backwater of a lowland reservoir

A comparative study of the floristic diversity of small rivers is of great importance in the assessment of their environmental state, which allows assessing the degree of pollution of the environment. The floristic diversity of the estuaries of the small rivers Korozhechna, Latka, Il’d’, and Chesnava, has been studied with special attention to the ecological groups and biological peculiarities of certain species. All the studied rivers flow into the Rybinsk Reservoir and have different morphology of the studied estuaries. The largest number of species has been recorded for the Chesnava River, the lowest, for the Il’d’ River. The representatives of families Poaceae, Cyperaceae, and Juncaceae evidence on the active overgrowing of shallow waters and periodically flooded coasts. Most of the species can grow on various soils, they are typical for water bodies with an oscillating water level and weak flow. The species-to-genus ratio, which is inversely proportional to the diversity of ecological conditions, is the highest in the Korozhechna River and the smallest in the Chesnava and Latka rivers. On the rivers Hydrophytes and hygrophytes dominated in the Korozhechna and Latka rivers; hygrophytes, hydrophytes, and hydrogelophytes, in the Il’d’ River; hygrophytes, mesophytes and hydrophytes, in the Korozhechna River. Such differences are explained by the peculiarities of environmental conditions. The highest similarity, by the absolute number of common species and by Jaccard coefficient, is found between the Chesnava River and the Latka River and between the Il’d’ River and the Latka River. According to hydrophytic cover index (HCI), near-water species dominated in the Chesnava River, a slight advantage in the proportion of the aquatic component of the flora was observed in the other rivers.

Observed evaporation dynamics from a large lowland reservoir during a hot summer

<p>In the past, most field studies on evaporation have focussed on land-atmosphere interactions, while the turbulent exchange above inland water surfaces have remained underexposed. However, due to the differences in characteristics of a land surface and a water body there are other driving mechanisms underlying the process of evaporation. This results in a difference in dynamics of surface evaporation between the land use types and consequently should lead to a different parameterization in hydrological models. Especially in a changing climate the importance of having an understanding of the driving mechanisms of open water evaporation (E­<sub>water</sub>) becomes more crucial to better predict to what extent the quantity and dynamics of E<sub>water</sub> could change in the future. This is essential to improve the parameterization of E<sub>water</sub> in operational hydrological models and therefore to optimize water management now and in the future. For this purpose, we set-up a long-term measurement campaign to measure E<sub>water</sub> and related meteorological variables over a large lowland reservoir in the Netherlands.</p><p>During the hot summer of 2019 two eddy-covariance systems were operational around lake IJsselmeer in the Netherlands. These high-temporal measurements are used to study the dynamics and to identify the forcing mechanisms of E<sub>water</sub>. We present the turbulent heat flux dynamics at several temporal scales over the summer season of 2019 and show how they are related to potential drivers and parameters. From this we develop a simple data based model for estimating hourly E<sub>water</sub> rates. Additionally, we compare E<sub>water</sub> resulting from the direct measurements to E<sub>water</sub> derived from commonly used evaporation models. Furthermore, we investigate and discuss the effect of including spatial variability on the total water loss of the IJsselmeer through E<sub>water</sub>. We achieve this by using the skin water temperature, which is considered an important predictor in the estimation of E<sub>water</sub>. Therefore, we use satellite products containing this information to extrapolate the in-situ observations towards spatially distributed rates of E<sub>water</sub>.</p>

Evaporation from a large lowland reservoir – (dis)agreement between evaporation models from hourly to decadal timescales

Accurate monitoring and prediction of surface evaporation become more crucial for adequate water management in a changing climate. Given the distinct differences between characteristics of a land surface and a water body, evaporation from water bodies requires a different parameterization in hydrological models. Here we compare six commonly used evaporation methods that are sensitive to different drivers of evaporation, brought about by a different choice of parameterization. We characterize the (dis)agreement between the methods at various temporal scales ranging from hourly to 10-yearly periods, and we evaluate how this reflects in differences in simulated water losses through evaporation of Lake IJssel in the Netherlands. At smaller timescales the methods correlate less (r=0.72) than at larger timescales (r=0.97). The disagreement at the hourly timescale results in distinct diurnal cycles of simulated evaporation for each method. Although the methods agree more at larger timescales (i.e. yearly and 10-yearly), there are still large differences in the projected evaporation trends, showing a positive trend to a more (i.e. Penman, De Bruin–Keijman, Makkink, and Hargreaves) or lesser extent (i.e. Granger–Hedstrom and FLake). The resulting discrepancy between the methods in simulated water losses of the Lake IJssel region due to evaporation ranges from −4 mm (Granger–Hedstrom) to −94 mm (Penman) between the methods. This difference emphasizes the importance and consequence of the evaporation method selection for water managers in their decision making.

Evaporation from a large lowland reservoir – (dis)agreement between evaporation methods at various timescales

Accurate monitoring and prediction of surface evaporation becomes more crucial for adequate water management in a changing climate. Given the distinct differences between characteristics of a land surface and a water body, evaporation from water bodies require a different parameterization in hydrological models. Here we compare six commonly used evaporation methods that are sensitive to different drivers of evaporation, brought about by a different choice of parameterization. We characterize the (dis)agreement between the methods at various temporal scales ranging from hourly to 10-yearly periods, and we evaluate how this reflects in differences in simulated water losses through evaporation of lake IJsselmeer in The Netherlands. At smaller timescales the methods correlate less (r = 0.72) than at larger timescales (r = 0.97). The disagreement at the hourly timescale results in distinct diurnal cycles of simulated evaporation for each method. Although the methods agree more at larger timescales (i.e. yearly and 10-yearly), there are still large differences in the projected evaporation trends, showing a positive trend to a more (i.e. Penman, De Bruin–Keijman, Makkink and Hargreaves) or lesser extent (i.e. Granger–Hedstrom and FLake). The resulting discrepancy between the methods in simulated water losses of the IJsselmeer region due to evaporation is ranging from −4 mm (Granger–Hedstrom) to −94 mm (Penman) between the methods. This difference emphasizes the importance and consequence of the evaporation method selection for water managers in their decision making.