Algae Blooms Disturb Lake Iroquois
Analyzing the Spread of Blue-Green Algae Using Remote Sensing Methods
Algae cover increased 23,318 sq. meters (½ football field) from 2012 to 2014
Across Vermont, blue-green algae blooms have been closing lakes due to their release of natural toxins. Beaches in Burlington were closed last summer due to blooms in Lake Champlain, and Lake Carmi up in Franklin County has remained closed for almost a month because of an intense bloom of more blue-green algae. The blue-green algae, scientifically known as cyanobacteria, contaminates waters with a blue-green slime and releases toxins upon decomposition that can cause health problems from skin rashes to diarrhea to those who come in contact with the murky waters.
Although Lake Iroquois has fortunately not seen bad algae blooms this summer, they have occurred in the past. This study examines the presence of cyanobacteria in Lake Iroquois using satellite imagery and remote sensing to track the growth of algae in the lake from 2012 to 2014. While blue-green algae levels in the lake were low in 2012, there was a increase in the area of algae in the lake in 2014 of 23,318 square meters, or ½ of a football field.
Video from Lake Carmi, VT in August 2017 of fish skimming the surface trying to breathe through the layer of blue-green algae slime. (YouTube)
Map comparing the algae growth on Lake Iroquois between 2012 and 2014. Popouts show the NAIP imagery for the northern end of the Lake, which demonstrated the highest levels of algae growth. (Nathan Beningson, NAIP)
Remote Sensing
Remote sensing is the study of Earth from a distance using sensors- think satellites or airplanes. It allows scientists to study inaccessible or large areas fairly quickly, and can provide a useful bird’s eye view of a region. While lots of remote sensing uses visible light and sensors resembling fancy cameras to gather data, other types of light such as infrared radiation or lasers can be used to image Earth’s surface too.
Other Uses of Remote Sensing
Mapping the retreat of Greenland's glaciers
(Moon & Joughin, 2008)
How does NDVI work?
The data used to look at Lake Iroquois came from the National Agricultural Imagery Program (NAIP). The NAIP is remotely sensed every three years using aircraft while vegetation is green and growing, and includes normal visible light (red, green, blue) and infrared light (heat energy). This “leaf-on” imagery useful for studying agriculture or plant growth, including algae.
In order to make the vegetation really pop out of the image and be easily identifiable, calculations can be run on the imagery that focus on the red and infrared light, creating an index called NDVI, or Normalized Difference Vegetation Index (see sidebar). When NDVI is used on the imagery of Lake Iroquois, the vegetation has a high NDVI while the lake and surrounding roads do not, making the vegetation surroudning the lake, as well as the algae growth on the lake more identifiable.
Methods and Results
Graph of NDVI along the cross section of the northern end of Lake Iroquois (see map). Higher values of NDVI correspond with the algae bloom in 2014.
To start, the imagery of the lake and surrounding area for 2012 and 2014 were analyzed using the NDVI index. The algae is visible from NDVI alone, as shown in the graph. Open water has very low NDVI values, while the high NDVI values correspond with areas covered in reflective cyanobacteria vegetation.
The data from 2012 and 2014 were then compared by finding the differences between the two NDVI datasets. Areas that did not change between the two years, like most of the surrounding roads, have consistently low NDVI values, while areas on the lake where algae has grown have a large difference in NDVI values.
The areas of difference were classified by the computer based on their various statistical attributes, which were then manually observed to find areas of algae growth in a process known as an unsupervised classification. The computer alone cannot identify where algae has appeared, but can divide the imagery into data-derived, computer-drawn categories. By looking to see which categories represent the desired areas of study, a human can quickly analyze large and detailed images without having to pore over every single pixel. The computer-grouped class suspected to contain algae were then isolated and used to create the map and calculate the increase in area on the lake covered by algae.
NAIP
NDVI
ALGAE
CLASSES
What do we do now?
Many cyanobacteria outbreaks can be tied to increased levels of phosphorous transport from farm runoff and warmer temperatures. Phosphorous-rich fertilizers runoff from agricultural fields into local streams, which is then transported into lakes, where the additional nutrients can promote unhealthy levels of algae growth. Much of the algae blooms in Vermont are attributed to agriculture runoff, especially from the dairy industry. Warmer summer temperatures strengthened by climate change can also promote algae growth.
Possible management solutions to target sources of excessive agricultural phosphorous and nutrient transport include better management of manure produced by livestock, decreased fertilizer use, or strengthening of riparian buffers along streams that help contain the spread of fertilizers into the water system (Sharpley et al., 2001).
Fortunately, conservation measures have shown to be successful- even on Vermont’s largest scale. Measures introduced to reduce agricultural phosphorus transport into Lake Champlain have been fairly successful, according to a recent report issued by the USDA Natural Resources Conservation Service (NRCS). Through effective management of agricultural runoff, Lake Iroquois residents can look forward to enjoying the pristine waters of the lake free from the blue-green cyanobacteria slime.
Works Cited
Friedman, R. A., Sofaer, A., & Weiner, R. S. (2017). Remote Sensing of Chaco Roads Revisited.
Moon, T., & Joughin, I. (2008). Changes in ice front position on Greenland's outlet glaciers from 1992 to 2007. Journal of Geophysical Research: Earth Surface, 113(F2).
Niethammer, U., James, M. R., Rothmund, S., Travelletti, J., & Joswig, M. (2012). UAV-based remote sensing of the Super-Sauze landslide: Evaluation and results. Engineering Geology, 128, 2-11.
Sharpley, A. N., McDowell, R. W., & Kleinman, P. J. (2001). Phosphorus loss from land to water: integrating agricultural and environmental management. Plant and soil, 237(2), 287-307.
Images
Friedman, R. A., Sofaer, A., & Weiner, R. S. (2017). Remote Sensing of Chaco Roads Revisited.
Moon, T., & Joughin, I. (2008). Changes in ice front position on Greenland's outlet glaciers from 1992 to 2007. Journal of Geophysical Research: Earth Surface, 113(F2).
Niethammer, U., James, M. R., Rothmund, S., Travelletti, J., & Joswig, M. (2012). UAV-based remote sensing of the Super-Sauze landslide: Evaluation and results. Engineering Geology, 128, 2-11.
Sharpley, A. N., McDowell, R. W., & Kleinman, P. J. (2001). Phosphorus loss from land to water: integrating agricultural and environmental management. Plant and soil, 237(2), 287-307.
Bibliography
http://www.healthvermont.gov/health-environment/recreational-water/cyanobacteria-blue-green-algae
https://vtdigger.org/2017/09/17/lake-carmi-closed-algae-bloom/#.Wb8b27pFxq1
http://www.mynbc5.com/article/all-burlington-beaches-open-as-blue-green-algae-blooms-clear/3327726
https://ethiopia-gis.nrel.colostate.edu/pdf/RSLessons/RS_Lesson_6_Spectral_Indices.pdf