The LAGOS-US Research Platform includes the 479,950 lakes and reservoirs > 1 ha in the 48 conterminous US states and tribes (Figure 1). LAGOS-US includes 3 core data modules (LOCUS, GEO, LIMNO) that are described below, an R package for easily accessing the core modules, GIS datafiles for all lakes and watersheds within the geographic extent, and the ArcGIS code to create many of the metrics in the LOCUS and GEO modules. Currently (as of August 2021) only some of the LAGOS-US Research Platform is available. Components of the platform will be made available as soon as possible. LAGOS-US also has additional extension modules (RESERVOIR, LAKE DEPTH, NETWORKS, LANDSAT) that the LAGOS team has created or is creating that are also described below.
Figure 1. Diagram of the LAGOS-US research platform, consisting of three core modules (GEO, LOCUS, LIMNO) and four extension modules (RESERVOIR, LAKE DEPTH, NETWORKS, LANDSAT).
Figure 2. The 479,950 lakes and reservoirs ≥ 1 ha in the conterminous US.
LAGOS-US Core Modules
LAGOS-US LOCUS (Available August 2021) Contains locational (i.e., latitude, longitude, elevation), identifying (i.e., lake names and identifiers across various lake datasets), and physical (i.e., area, perimeter, shape) information of all lakes ≥ 1 ha and their watersheds in the conterminous U.S. Additionally, LOCUS provides a dynamic classification of freshwater connectivity (e.g., lakes that are hydrologically isolated seasonally) and identifies all lakes that were previously glaciated. Geospatial files for lakes and watersheds as well as tables for linking across other LAGOS and broad-scale lake databases are available. All LOCUS data can be linked across LAGOS data products using the unique lake identifier (lagoslakeid). A data paper that presents this important data module more fully has been published (Cheruvelil et al. 2021).
LAGOS-US GEO (Expected availability late 2021) contains geospatial and temporal ecological context variables for all lakes in LOCUS characterized at multiple spatial divisions (e.g., equidistant buffers around lakes, watersheds, ecoregions). Ecological context variables were derived from various public datasets and pertain to land use/land cover, climate, hydrology, freshwater connectivity, terrain, soils, land ownership and protection, atmospheric pollution, and other human activities such as roads and dams.
LAGOS-US LIMNO (Expected availability summer 2022) contains in situ surface-water and profile limnological physical, chemical, and biological measurements for a subset of lakes ≥ 1 ha through time. Water quality data were collected from the late 1980s to 2017 by approximately 250 different state, federal, citizen science, nonprofit, tribal, and university lake monitoring programs. We harmonized all data to ensure data compatibility across all programs (e.g., we standardized units and checked lab and field methods). In addition to the common water quality variables found in LAGOS-NE-LIMNO such as water clarity, nutrients, chlorophyll-a, dissolved organic carbon, and color, LAGOS-US-LIMNO provides observations of water temperature, dissolved oxygen, pH, alkalinity, conductivity, numerous ions and contaminants, turbidity, and microcystin. Detailed information on sampling (i.e., methods, dates, depths), as well as laboratory methods, are provided alongside lake observations.
LAGOS-US Extension Modules
LAGOS-US RESERVOIR (Expected availability late 2021) provides a predicted classification of all 137,465 lakes ≥ 4 ha as either a natural lake or reservoir using a machine-learning algorithm and aerial imagery (Figure 3). Classification accuracy was approximately 80%.
Figure 3. Natural lakes (NL) and reservoirs (RSVR) (≥ 4 ha) in the conterminous US classified in LAGOS-US-RESERVOIR.
LAGOS-US LAKE DEPTH (Expected availability September 2021) represents a manual compilation of mean and/or maximum depth of over 17,675 lakes ≥ 1 ha from a wide range of online sources (Figure 4).
Figure 4. Lakes in the conterminous US with depth data in LAGOS-US-DEPTH.
LAGOS-US NETWORKS (Available August 2021) consists of 898 lake networks (86,511 total lakes ≥ 1ha) and provides quantitative freshwater connectivity metrics for those networks and lakes (Figure 5). A data paper that further explores this data module has been published (King et al. 2021)
Figure 5. The 898 lake networks in the conterminous US in LAGOS-US-NETWORKS. Individual colors represent unique networks.
LAGOS-US-LANDSAT (Expected available in 2022) provides predicted water quality measurements for chlorophyll-a, water clarity, and colored dissolved organic matter for all lakes ≥ 4 ha. LANDSAT uses machine-learning models based on atmospherically corrected Landsat imagery and LIMNO data, in addition to lakewide values of reflectance for each Landsat band and satellite overpass.