Climate change is affecting lake stratification with consequences for water quality and the benefits that lakes provide to society. Here we use long‐term temperature data (1970–2010) from 26 lakes around the world to show that climate change has altered lake stratification globally and that the magnitudes of lake stratification changes are primarily controlled by lake morphometry (mean depth, surface area, and volume) and mean lake temperature. Deep lakes and lakes with high average temperatures have experienced the largest changes in lake stratification even though their surface temperatures tend to be warming more slowly. These results confirm that the nonlinear relationship between water density and water temperature and the strong dependence of lake stratification on lake morphometry makes lake temperature trends relatively poor predictors of lake stratification trends.

Climate warming is expected to have large effects on ecosystems in part due to the temperature dependence of metabolism. The responses of metabolic rates to climate warming may be greatest in the tropics and at low elevations because mean temperatures are warmer there and metabolic rates respond exponentially to temperature (with exponents >1). However, if warming rates are sufficiently fast in higher latitude/elevation lakes, metabolic rate responses to warming may still be greater there even though metabolic rates respond exponentially to temperature. Thus, a wide range of global patterns in the magnitude of metabolic rate responses to warming could emerge depending on global patterns of temperature and warming rates. Here we use the Boltzmann-Arrhenius equation, published estimates of activation energy, and time series of temperature from 271 lakes to estimate long-term (1970-2010) changes in 64 metabolic processes in lakes. The estimated responses of metabolic processes to warming were usually greatest in tropical/low-elevation lakes even though surface temperatures in higher latitude/elevation lakes are warming faster. However, when the thermal sensitivity of a metabolic process is especially weak, higher latitude/elevation lakes had larger responses to warming in parallel with warming rates. Our results show that the sensitivity of a given response to temperature (as described by its activation energy) provides a simple heuristic for predicting whether tropical/low-elevation lakes will have larger or smaller metabolic responses to warming than higher latitude/elevation lakes. Overall, we conclude that the direct metabolic consequences of lake warming are likely to be felt most strongly at low latitudes and low elevations where metabolism-linked ecosystem services may be most affected.

Globally, lake surface water temperatures have warmed rapidly relative to air temperatures, but changes in deepwater temperatures and vertical thermal structure are still largely unknown. We have compiled the most comprehensive data set to date of long-term (1970–2009) summertime vertical temperature profiles in lakes across the world to examine trends and drivers of whole-lake vertical thermal structure. We found significant increases in surface water temperatures across lakes at an average rate of + 0.37 °C decade−1, comparable to changes reported previously for other lakes, and similarly consistent trends of increasing water column stability (+ 0.08 kg m−3 decade−1). In contrast, however, deepwater temperature trends showed little change on average (+ 0.06 °C decade−1), but had high variability across lakes, with trends in individual lakes ranging from − 0.68 °C decade−1 to + 0.65 °C decade−1. The variability in deepwater temperature trends was not explained by trends in either surface water temperatures or thermal stability within lakes, and only 8.4% was explained by lake thermal region or local lake characteristics in a random forest analysis. These findings suggest that external drivers beyond our tested lake characteristics are important in explaining long-term trends in thermal structure, such as local to regional climate patterns or additional external anthropogenic influences.

a b s t r a c tLake Atitlan, one of the most important lakes not only in Central America but in the whole world, is facing serious problems with increasing water pollution. Over the last several decades, the uncontrolled nutrient input into the lake has lead to high P levels and low N:P ratios, initiating cyanobacterial blooms. The first bloom occurred in December of 2008, followed by more extensive bloom in October 2009. The blooms are formed by cyanobacteria from the rare planktic Lyngbya hieronymusii/birgei/robusta complex. Based on the species morphology, the Atitlan population corresponds to L. robusta and this is the first case of reported bloom of this species worldwide. Remote sensing images documented that at the maximum bloom development, 40% of the 137 km 2 of the lake area were covered by dense patches of Lyngbya, with the chlorophyll a concentration reaching over 100 g L −1 . The only toxins detected in the 2009 bloom were trace levels of cylindrospermopsin and saxitoxin with 12 and 58 ng g −1 , respectively. The nitrogen fixation followed a pattern expected in non-heterocytous cyanobacteria, i.e., the nitrogenase activity was minimal during the day, while during the night the activity reached 2.2 nmol C 2 H 4 g Ch a −1 h −1 . Delta 15 N of −0.86‰ was well in the range given for nitrogen fixing organisms. The cell C, N and P content was 36.7%, 5.9% and 0.9%, respectively, resulting in the molar ratio of 105:14.4:1. A well designed and executed lake monitoring program, strict control of nutrient input into the lake, and public education are the necessary prerequisites for potential prevention of even more severe blooms than the one from 2009.

In this study we evaluated the applicability of a space-borne hyperspectral sensor, Hyperion, to resolve for chlorophyll a (Chl a) concentration in Lake Atitlan, a tropical mountain lake in Guatemala. In situ water quality samples of Chl a concentration were collected and correlated with water surface reflectance derived from Hyperion images, to develop a semi-empirical algorithm. Existing operational algorithms were tested and the continuous bands of Hyperion were evaluated in an iterative manner. A third order polynomial regression provided a good fit to model Chl a. The final algorithm uses a blue (467 nm) to green (559 nm) band ratio to successfully model Chl a concentrations in Lake Atitlán during the dry season, with a relative error of 33%. This analysis confirmed the suitability of hyperspetral-imagers like Hyperion, to model Chl a concentrations in Lake Atitlán. This study also highlights the need to test and update this algorithm with operational multispectral sensors such as Landsat and Sentinel-2.