axe
] ena-
lake
sed
to the
ation
ate in
attem-
e Mul-
in Hy-
luded
ertan-
devia-
an re-
e ob-
e re-
para-
les of
ion
alysis
produ-
the
the
re-
ge of
sets
The
the
he
the
paren-
ity,
etc.
dica-
rea
l ef-
adual
int to
xe
ll in-
ing
noe
other
of
ERES a a S ue
growth of water weeds (Eichhornia Crassepis) with more than 50 per
cent of surface area covered by the plants. Unsuitable foreshore
practices such as the large park under development, inflow of se-
wage and human waste from nearby colonies, the unmecadamed road
jutting into the lake compound the pollution problem, in addition
to the Kukatpalli drain bringing industrial wastes into the lake.
The water weeds have also contributed to an increase in the mos-
quito population leading to a general deteioration in civic health
level. t
lh. CONCLUSIONS AND RECOMMENDATIONS
A major problem in remote sensing is that the signal reach-
ing the sensor may represent a mix of colour, bottom reflectance
and turbidity produced by different suspended material. such as al-
gae and sediments. Changes in the relative amounts of constituents
may occur over time as well as changes in the size, shape and
colour of suspended materials. Hence in monitoring load concentra-
tions, eutrophication status or dissolved solids, it is necessary
to determine the reasons for change in water turbidity or colour.
Thecombined effect of turbidity and colour on the spectral signa-
tures of water can be quite complex and need to be studied in de-
tail before attempting water quality mapping of any specific agua-
tic environment. In this study, fortun-tely, the colour of Husaïn-
sagar lake was uniform and also the turbidity vas mainly caused by
algae. In Godavari river the turbidity, which was low, is mainly
caused by silt carried by the river, and the significant charac-
teristic of the paper mill effluent when it enters the river water
is its colour and not turbidity, due to settling in the shallow
lagoons. Thus the conflicting colour-turbidity interaction on the
spectral response has been avoided.
Water sample collection is an important component of the re-
mote sensing process. Proper sampling strategies by way of collec-
tion of adequate number of samples at suitable sites, correct pre-
servation and storage procedures and proper analysis methods con-
tribute greatly to the succes: of any remote sensing project. The
logistics of sampling effort need to be carefully planned and exe-
cuted.
The dynamic waters of Godavari river necessitated simulta-
neous sampling while time-lag between aircraft over-pass and
sample collection could be permitted in Husainsagar lake.
The aerial flights need to be properly planned in order to
acquire remotely sensed data suitable for water quality studies.
Season of the year, time of day, aircraft oltitude, flight direc-
tion and line spacing are some of the major mission parameters to
be considered. The mission planning should also take into account
limitations imposed by sampling requirements.
Though in this study water quality parameters have been in-
dividually related to spectral response and mapped, it is prefera-
ble to group various parameters into indices which will have more
physically meaningful and statistically significant relationships
to spectral response. The grouping of parameters may be performed
through principal components analysis. The first one or two prin-
cipal components may be adequate to represent the water quality
615
M
mem RUNDE Due Li e RA ER EN M
