20 Remote Sensing of African Lakes: A Review
415
Region experienced increases in estimated chlorophyll two months after the onset
of lake surface warming. Differential cooling in the north and south were shown to
drive the large-scale convective circulation observed in the lake, a pattern that offers
a possible explanation for the observed bio-optical patterns. This study revealed how
year-to-year variability in physical conditions may play a major role in phytoplankton
dynamics, irrespective of the influence of nutrient dynamics.
Matthews et al. (2012) developed a novel algorithm for detecting chlorophyll-a,
cyanobacterial blooms, surface scum and floating vegetation using MERIS imagery
of three diverse African lakes (Loskop dam, Hartbeespoort dam, Zeekoevlei Lake)
and coastal ocean (southern Benguela). The Maximum Peak Height (MPH) algorithm
uses the top-of-atmosphere MERIS spectral data, not the atmospherically corrected
data commonly used by bio-optical algorithms targeting inland water bodies.
The MPH is designed to provide accurate chlorophyll estimation in three cases:
(1) mixed oligotrophic/mesotrophic low-medium biomass waters, (2) high biomass
eutrophic/hypereutrophic waters and (3) extremely high biomass waters with surface
scums or floating vegetation. The parameterization of the algorithm for each potential
peak provided reasonable chlorophyll estimates across a vast range of trophic states
(mean absolute percentage error = 59.9 %, r
2
= 0.71), providing a promising new
tool for remote sensing of lakes throughout Africa and beyond.
20.6 Discussion
Remote sensing offers the potential to provide a rich body of data for lake management. Whether the concern about a lake relates to area, volume, temperature or
trophic state, remote sensing has shown to be appropriate for use in African lakes.
In the studies covered here, a large array of sensors (17) have been used to measure
features of 29 African lakes, providing technical developments, real application, or
both.
While these 29 lakes represent only a fraction of the more than 17,000 lakes
in Africa identified in the Shuttle Radar Topography Mission (SRTM) Waterbody
Dataset (SWBD) of NASA (NASA 2013), they cover 76 % (186,192 km
2 ) of the total
lake area (245,137 km
2 ) in Africa. In addition, 13 of the largest 15 lakes in terms of
area are included. These larger lakes are those for which remote sensing can be most
useful, providing information on a spatial and temporal scale simply not possible with
traditional sampling methods. As physical and biological information becomes more
available through remote monitoring and analysis, a better understanding of lake
conditions and lake dynamics will emerge, paving the way for better management of
these important resources. It should also be noted that many applications of remote
sensing are used for internal policy documents and unpublished land cover analyses
and do not make it into the literature. These practical, informal uses of remote sensing
are often not fully appreciated.
415
Region experienced increases in estimated chlorophyll two months after the onset
of lake surface warming. Differential cooling in the north and south were shown to
drive the large-scale convective circulation observed in the lake, a pattern that offers
a possible explanation for the observed bio-optical patterns. This study revealed how
year-to-year variability in physical conditions may play a major role in phytoplankton
dynamics, irrespective of the influence of nutrient dynamics.
Matthews et al. (2012) developed a novel algorithm for detecting chlorophyll-a,
cyanobacterial blooms, surface scum and floating vegetation using MERIS imagery
of three diverse African lakes (Loskop dam, Hartbeespoort dam, Zeekoevlei Lake)
and coastal ocean (southern Benguela). The Maximum Peak Height (MPH) algorithm
uses the top-of-atmosphere MERIS spectral data, not the atmospherically corrected
data commonly used by bio-optical algorithms targeting inland water bodies.
The MPH is designed to provide accurate chlorophyll estimation in three cases:
(1) mixed oligotrophic/mesotrophic low-medium biomass waters, (2) high biomass
eutrophic/hypereutrophic waters and (3) extremely high biomass waters with surface
scums or floating vegetation. The parameterization of the algorithm for each potential
peak provided reasonable chlorophyll estimates across a vast range of trophic states
(mean absolute percentage error = 59.9 %, r
2
= 0.71), providing a promising new
tool for remote sensing of lakes throughout Africa and beyond.
20.6 Discussion
Remote sensing offers the potential to provide a rich body of data for lake management. Whether the concern about a lake relates to area, volume, temperature or
trophic state, remote sensing has shown to be appropriate for use in African lakes.
In the studies covered here, a large array of sensors (17) have been used to measure
features of 29 African lakes, providing technical developments, real application, or
both.
While these 29 lakes represent only a fraction of the more than 17,000 lakes
in Africa identified in the Shuttle Radar Topography Mission (SRTM) Waterbody
Dataset (SWBD) of NASA (NASA 2013), they cover 76 % (186,192 km
2 ) of the total
lake area (245,137 km
2 ) in Africa. In addition, 13 of the largest 15 lakes in terms of
area are included. These larger lakes are those for which remote sensing can be most
useful, providing information on a spatial and temporal scale simply not possible with
traditional sampling methods. As physical and biological information becomes more
available through remote monitoring and analysis, a better understanding of lake
conditions and lake dynamics will emerge, paving the way for better management of
these important resources. It should also be noted that many applications of remote
sensing are used for internal policy documents and unpublished land cover analyses
and do not make it into the literature. These practical, informal uses of remote sensing
are often not fully appreciated.