| |
In-Situ and Remote Water Quality Monitoring
Importance of Water Quality Monitoring
Detection and identification of algal species in local water supplies provides information
about, and advance warning of, potentially harmful species present. Statistics of indicators
of the health of detected algal species may also portend other unseen issues in the ecological
surroundings. Automated classification to the taxa level would allow for more timely identification
of harmful species, as illustrated in Figure 1: a. Karenia brevis, b. Pfisteria sp., c. Prorocentrum
minimum, d. Chattonella verruculosa, among others, as well as establish their representative ratios
with respect to beneficial algal species.
Occasionally, just the right conditions exist that allow some species to grow rapidly,
producing harmful toxins or consuming natural resources to a degree that they impact the
immediate environment. These harmful algal blooms (HABs) may have serious impacts:
- Economic - a blow to commercial fishing, recreation, and tourism.
- Ecological - wildlife kills from HAB toxins, competition with beneficial algae, oxygen depletion in water.
- Human health - toxicity-related illnesses from natural or introduced harmful species.
Michigan Aerospace Corporation's Role
Michigan Aerospace Corporation has the background and expertise to combine automated water sampling
instruments and robust automated classification software to provide early detection of Harmful
Algal Blooms (HAB's). Michigan Aerospace is also experienced in remote sensing, which may be
applied to remote species detection based on species-specific fluorescence spectra.
In-situ Sampling
Michigan Aerospace's Data Exploitation Group has scientists experienced in pattern recognition and
automated learning algorithms, which are directly applicable to the task of automated sampling and
classification.
 |
In-situ water monitoring to identify and evaluate aquatic
species - for early HAB warning, or general environmental health
|
|
|
Automated recognition of species is achieved using a
powerful machine-learning paradigm. Automation is maintained from the start of sampling to
the end product - the reporting.
|
|
|
|
Contact Us 
b. 
c. 
d. 
