Lakes act as slow integrators of atmospheric disturbance like oceans: Evidence from a deep perialpline lake

Abstract

Slow and long-term variations of sea surface temperature anomalies have been interpreted as a red-noise response of the ocean surface mixed layer to fast and random atmospheric perturbations. How fast the atmospheric noise is damped depends on the mixed layer depth. In this contribution we provide first evidence that lakes are integrators of noisy atmospheric variability just like oceans are. Based on a stochastic approach inspired by the stochastic climate models theory by the 2021 Nobel Physics laureate Hasselmann, we determine estimates of surface mixed layer depth from satellite measurements of Lake Surface Water Temperature (LSWT). The proposed approach is showcased for Lake Garda, Italy. We demonstrate that LSWT anomalies have a red noise spectrum resulting from the integration of higher frequency atmospheric forcing. By connecting the decorrelation time scale of LSWT anomalies to net heat fluxes, we obtain a spatially varying estimate of mixed layer depth. The basin-scale variability of our estimate is consistent with in-situ measurements and connects to the dominant modes of LSWT and chlorophyll-a concentrations obtained via empirical orthogonal functions. We thus show that (i) remotely-sensed quantities also carry information on the relevant spatial and temporal scales of mixed-layer processes and (ii) there is a limit to the persistence, hence the predictability, of the anomalies of LSWT, which poses a physical constraint to temporal gap-filling procedures. The lessons learnt from ocean modelling is that such first-order picture necessarily overlooks finer scale dynamics, e.g. the effect of intense currents advecting water temperature vertically and horizontally, seasonal modulations and higher order modes of variability, which can be well described by more complex deterministic models. For such a reason, applications to spatial scales different than single points or small portions of the ocean are not common in marine literature. That kind of dynamics also affect lakes surface mixed layer, where spatial and temporal scales of thermal inertia shrink. Our study demonstrates that such a stochastic approach, rather classical in ocean literature, can be applied to the entire surface of enclosed basin and provides useful insights on the thermal and ecological heterogeneity beneath surface.