Relevance of inflows on the thermodynamic structure and on the modeling of a deep subalpine lake (Lake Maggiore, Northern Italy/Southern Switzerland)

•An enclosed 1D thermodynamic model of a subalpine lake gives satisfactory results.•An unrealistically low light extinction coefficient is used to heat lower layers.•Simulating the intrusion of main inflows gives better results over the thermocline.•Bottom hypolimnion dynamics cannot be precisely reproduced by 1D models in any case.•Major riverine interflows are crucial for heating the deep layers of Lake Maggiore.

Atmospheric exchanges largely dominate the heat budget of deep lakes in temperate regions. Heat import and export by through-flows is of much lower entity and has been neglected or simplified in many numerical thermodynamic models of lakes. This is often due either to the unavailability of data for inflows and outflows, or to the difficulties in forecasting the evolution of their discharge and temperature in climate change studies. While disregarding through-flows may seem correct, riverine intrusions can bring warmer water than the deep hypolimnetic one to the lower metalimnion and upper hypolimnion, where sunlight does not penetrate and mixing is poor. For holomictic lakes with significant inflow contributions, this can affect the thermal structure at intermediate depths, hampering any numerical model which neglects through-flows. This study focuses on a relevant basin under such aspect, Lake Maggiore (Northern Italy/Southern Switzerland), which drains the rainiest watershed of the Southern Alps. First, we quantify to what extent a one-dimensional fixed-level model ignoring through-flows is able to predict the observed evolution of the thermal structure of the lake and the improvements resulting from reproducing the main inflows and outflows. Then, we directly discuss the influence of through-flows on the thermodynamic structure of Lake Maggiore. The General Lake Model (GLM) was here adopted, reproducing the 1998-2014 period, spanning years with different meteorological and hydrological features. Results show that a calibrated enclosed-lake model can give satisfactory results only if it employs an unrealistically low light extinction coefficient to allow heating of the deep metalimnion and hypolimnion, whose real warming strongly depends on interflows.