Welcome to our harmful algal bloom forecasting dashboard. Here we use and show data collected from USGS streamgages, weather stations, satellite imagery, and on-the-ground water samples. These data are used by custom machine learning algorithms to forecast when and where harmful algal blooms will occur.

ClearWater Analytica is a public benefit company whose mission is to improve the resilience of communities to climate change driven water shocks. We synthesize data from a multitude of sources, and use custom machine learning algorithms to forecast changes in the quality of the water that we drink. More information about ClearWater Analytica can be found here.

Environmental data is collected by the The Climatology Lab through GRIDMET . This data set features daily high-spatial resolution (~4km, 1/24 degree) from 1979 - yesterday. The data available includes maximum temperature, minimum temperature, precipitation accumulation, downward surface shortwave radiation, wind-velocity, humidity (maximum and minimum relative humidity and specific humidity.

Sonde data collected by the City of Salem via handheld instrumentation and a YSI vertical profiler, deployed on a buoy near the Detroit Dam.

The CyAN tab provides results from our machine learning model that classifies whether CyAN was detected for a given week. The model forecasts the probability that CyAN will be detected in the next week using environmental data and site specific characteristics. The left donut chart provides the results from the previous week split between the probability of detect (aqua color pie) and probability of no detect (blue color pie). The percentage observed detect area is provided in the center. Next week's results are provided similar in the right donut with the model accuracy provided in the center. The line chart below provides the historical expected probability of detected CyAN (red line) with the current year's running probability overlayed (blue line). Clicking anywhere on the blue line will provide you with that weeks probability.

Satellite data is provided by a Multispectral Instrument (MSI) that is available on the European Union’s Copernicus Sentinel-2. The satellite launched in 2017 to provide high-resolution, multispectral images with a high revisit frequency for land and water features. Currently, data for 13 spectral bands at mid-latitudes is available every 2-3 days. The bands measure central wavelengths from 442.7-2185.7 nanometers (nm) which allows for identification of color bands. We scrape blue, green, red, near-infrared (NIR), water vapor, cloud mask, and snow probability for all available data from 2017-03-28 to present minus 3 days.

Additional satellite data is provided to describe potential cyanobacteria observations through the Cyanobacteria Assessment Network (CyAN) . CyAN data is provided by the Environmental Protection Agency (EPA), the National Aeronautics and Space Administration (NASA), the National Oceanic and Atmospheric Administration (NOAA), and the United States Geological Survey (USGS). This data provides cyanobacteria estimates for U.S. lakes and estuaries that can detect and quantify cyanobacteria algal blooms.

Red colored values highlight areas of high chlorophyll-a or CyAN, relative to the blue colored areas. Values are computed as the ratio of the Near Infrared and Red bands for chlorophyll-a from the Sentinel 2a satellite.