What is it about?
Scientists studied 24,452 U.S. lakes over 34 years to see if climate causes increases in algae, as many people expect. They found that climate does affect algae in about one-third of the lakes, but big, long-term increases are rare. Surprisingly, the lakes that show the strongest connections to climate are the ones that are least affected by human activities.
Featured Image
Photo by Matthew Daniels on Unsplash
Why is it important?
Understanding how climate change may alter the natural world is difficult to predict because different locations can respond differently, and climate doesn't change the same everywhere. These challenges can be addressed by taking a big-data approach using satellite images that can 'sample' everywhere. Our study demonstrates the value of this type of approach and shows where climate change is likely to be felt strongest in lakes across the US.
Perspectives
This study has been important for me because it demonstrates an approach to studying a lot of significant problems facing the natural world within the US and across the globe. Too often, we study problems at a local scale, where we can go out and sample ourselves and delve deeply into one site or ecosystem. Such studies are critical, but we do not always know how to extend the results to other sites or ecosystems. Our study presents a way to study thousands of ecosystems from the last 30+ years so that we can learn from all of the varied natural settings across the US, which gives us insight into where climate change may be most damaging.
Patricia Soranno
Michigan State University
Read the Original
This page is a summary of: Abrupt changes in algal biomass of thousands of US lakes are related to climate and are more likely in low-disturbance watersheds, Proceedings of the National Academy of Sciences, February 2025, Proceedings of the National Academy of Sciences,
DOI: 10.1073/pnas.2416172122.
You can read the full text:
Resources
Dataset for: Abrupt changes in algal biomass of thousands of US lakes are related to climate and are more likely in low-disturbance watersheds
These datasets include both the raw data and the model results from a project exploring causal relationships between climate and lake productivity of a diverse set of 24,452 US lakes over 34 years (1984-2008). Climate change is expected to increase lake productivity and algal blooms and cause regime shifts, particularly in human-impacted ecosystems. First we assessed the potential for nonlinear dynamics in summer median lake chlorophyll (CHL; derived from remote sensing) to demonstrate the potential for regime shifts in these lakes; second, we modeled the causal effects of climate on lake CHL over the 34 years. Climate was causally related to CHL in 34% of lakes; of those, 71% exhibited abrupt shifts, but only 13% had the potential for regime shifts. Climate-influenced lakes had low productivity at environmental extremes and experienced some directional climate change. Our synthesis of 24,452 time series demonstrates that effects of climate on lake productivity depend on recent climate change interacting with human impact and environmental context.
LAGOS-US LANDSAT: Data module of remotely-sensed water quality estimates for U.S. lakes over 4 ha from 1984 to 2020
This data package, LAGOS-US LANDSAT, is one of the extension data modules of the LAGOS-US platform that provides six water quality estimates (chlorophyll, Secchi depth, dissolved organic carbon, total suspended solids, turbidity, and true water color) from remote sensing for lakes ≥ 4 ha in the conterminous U.S. (48 states plus the District of Columbia) for the years 1984-2020. These estimates are generated through machine learning models on in-lake water quality matchups from LAGOS-US LIMNO with Landsat 5, 7, and 8 whole lake median reflectance values and pixel-wise band ratios that are subsequently used to make predictions across the U.S. The LANDSAT module contains remotely sensed reflectance values for 136,977 of the 137,465 lakes ≥ 4 ha from the LAGOS-US research platform. Within the module are a total of 45,867,023 sets of reflectance values, a matchup dataset with a window of up to 7 calendar days with in situ data, and associated water quality parameter predictions for each reflectance set. Additional quality control flags are provided for predictions indicating whether reflectance extractions included negative values, the percent of the maximum pixels ever retrieved for that lake that the predictions are based on, and whether there are shared calendar day predictions due to scene overlap.
Contributors
The following have contributed to this page
