Great Lakes Region

NOAA’s Great Lakes Environmental Research Laboratory (GLERL), in collaboration with the Cooperative Institute for Great Lakes Research (CIGLR) and Skypics, has been using an airplane mounted with a specialized piece of equipment on a camera, called a hyperspectral sensor.  This sensor captures high resolution images to improve harmful algal bloom (HAB) forecasts. Like satellites, hyperspectral cameras collect information on HAB location and size, but are more accurate and provide detailed information on bloom concentration, extent (or spread), and even the types of algae present in the lake. Hyperspectral sensors measure many more color wavelengths — bands of color in the visible light spectrum — than satellites, and can detect color pigments that represent different types of algal groups. This pigment detection information helps determine what type of algae is present within blooms and whether or not toxins are present.

The airplane flies weekly in western Lake Erie, and is coordinated with the weekly Lake Erie monitoring program, which helps assess how well the sensor is working for HAB detection. The monitoring program collects samples at multiple stations around the western part of the lake, and the hyperspectral sensor captures images from the sampling stations on the same day. Since the weekly airplane hyperspectral flyovers are done below the clouds, they are able to take images even when satellites cannot, and they obtain images that are much higher resolution compared to satellites.

There were many instances in 2016 where the camera was able to detect important algal scum — a thick algae layer on the surface of the lake — under clouds or near the shoreline, which satellites are unable to do. A prime example was on September 19, 2016 when algal scums were detected in nearshore areas with greater frequency, and also on a follow up flyover on September 22, 2016, when scientists detected algal scums under cloud features. The hyperspectral camera images were converted to the cyanobacteria index (CI), which is similar to the satellite algorithm used to develop the weekly NOAA Lake Erie bulletin.  These were overlaid on the NOAA Lake Erie HAB bulletin image using the same color table.  By using hyperspectral imagery in conjunction with the satellite imagery, scientists are able to provide more robust information to municipal water managers and decision-makers, which better supports the health of our ecosystems and drinking water.

Image of Hyperspectral Satellite

a) The NOAA Lake Erie HAB bulletin for 09-18-2016, one day before the hyperspectral flyover on 09-19-2016, showing the satellite with a red box designating where the bulletin missed a scum feature near the shore. b) The hyperspectral camera algae image. The same color table and algorithm as the bulletin has been applied to the hyperspectral imagery. c) A true-color comparison from the hyperspectral sensor of the same images of the scum feature (shown in green).

Below are three hyperspectral camera images chosen for their representation of the algae features that persisted north of the islands in the western basin of Lake Erie on September 22, 2016. The spectral resolution of each pixel is approximately 1 meter, much smaller and higher resolution than a typical satellite that has 1 km pixels. Hyperspectral cameras have the unique capability to capture the spatial variability of scums. One image is about the same size as one pixel that a satellite sees.

Hyperspectral imagery

STAY CONNECTED
NOAA Facebook NOAA Twitter RSS Feed for this site Email Us NOAA's Flickr Feed