Community science helps digitize 78 years of fish and habitat data for thousands of lakes in Michigan, USA.

North temperate lakes are an important resource across North America and Eurasia, however, their ecosystems are declining and projected to continue to face further impacts under future land use and climate change. Understanding how lake ecosystems respond to environmental stressors and management actions is critical for identifying resilient lakes and developing adaptation strategies. However, the ability to manage lakes is hampered by a lack of historical information. We describe our methods to produce a usable (i.e. machine-readable, uniform, and standardized) historical dataset of inland lake habitat and fish communities across the State of Michigan, United States of America. Historical data were originally archived as index cards at the Michigan Department of Natural Resources. Using our workflows, the cards were transcribed by community scientists and subsequently curated. We focused on three types of data records: lake summary cards (lake characteristics and fish species present), fish collection cards (gear, effort, and fish species counts), and fish growth cards (species length-at-age) from >1,300 lakes across 78 years (1921-1998).