Abstract
Urban lakes are heavily impacted by human activities and climate variability, and they provide many ecosystem services to residents. The MSP LTER program is studying long term changes in urban lake water quality, ecology and management as part of our long term studies of urban environments. The goal of this dataset is to understand how land-use change, management, and climate have impacted urban lake biogeochemistry over time. This dataset includes parameters characterizing the long term (> 5 years) surface water quality and chemistry of 294 lakes and ponds in the Minneapolis-Saint Paul Seven County Metropolitan Area, Minnesota, USA. The dataset draws from data publicly available through the Minnesota Pollution Control Agency and data provided by individual agencies, park districts and cities. The dataset is distinct from other lake datasets because it is curated to only report a single value per lake x date x parameter, minimizing the amount of data manipulation needed before use in statistical analyses. All data come from the top two meters of the water column. In the case of multiple spatial measurements on a single lake or multiple agencies sampling the same lake on the same day, chemistry data were averaged to generate a single value. For Secchi data, the deepest reported observation on a given lake x date was used. Parameters: total phosphorus, total nitrogen, total Kjeldahl nitrogen, nitrate, nitrite, nitrate + nitrite (NOx), ammonium, chlorophyll a (corrected and not corrected for pheophytin), specific conductivity, chloride, and Secchi depth. These waterbodies are identified by their DNR Division of Water (DOW) number with minor alterations for subbasin identification. This dataset does not comprehensively represent all lentic waterbodies that have substantial water quality data in the metro area, and some included waterbodies may be considered wetlands according to state classifications. The data brought together in this database has undergone QAQC by the organizations that originally collected it, as well as a screening process during data harmonization. While we believe that the resulting dataset is robust, we cannot guarantee that it is free of errors or inaccuracies.