Pollutant Diffusion in
Loxahatchee Inlet
(Influence of Physical Environment on Pollutant Concentration with GIS)
MingYo Chung
Introduction
Until recently, emphasis by environmental engineers and decision makers
had been directed toward the treatment of traditional point source
pollution. Billions of dollars were spent in the United States on the
point source pollutant cleanup mandated by the earlier version of the
Clean Water Act (1972, 1977). As a result of these policies, marked
improvements of water quality of some bodies of water were noticed such
as the River Thames in London, and Potomac River at Washington, D.C.
However, Wolman (1988) focused on the point source abatement has only
maintained more or less a status quo in the majority of water quality
monitoring stations throughout the United States. It has to be pointed
out, that most of the monitoring stations operated by the U.S.
Geological Survey may not be located in places where the most profound
water quality changes have occurred. Hence the recent experience of
emphasis on the point source abatement indicates that focusing on one
type of pollution may not be efficient and that an integrated approach
that would address both point and non point sources is
needed.
The elimination or reduction of pollution sources may require excessive
expenses. Furthermore, many diffuse (non point) sources cannot be
regulated and enforcement of control is not feasible. Therefore, the
integrated solutions or management plan should address the magnitude of
the water pollutant sources. This Loxahatchee Inlet data shows both
point and non point source pollution, and this study focused on
visualizing pollutant concentration with surrounding environment
throughout Loxahatchee Inlet.
Purpose
Three fundamental purposes are involved in this study.
1). Find spatial relationship between pollutants concentration and
main boat path, depth, bottom type
2).
Try to explain the pattern of pollutants concentration in terms of
other features of the inlet
3).
Help to developing integrated management modeling and
plans
Data Description
Loxahatchee Data include:
Shorelines: determine the study area defined (polygon)
Depth: shows entire depth (ft) in the Loxahatchee Inlet (polygon)
Sample station with pollutants concentration: represent concentration of
each
point (ppm) ? (point)
Bottom type: shows fine different bottom type in the Loxahatchee Inlet
(polygon)
Main
boat path: heavy boat traffic (line)
Summerhouses location: point source pollution (point)
Method
Quadrat Analysis:
Defining spatial randomness of the sample station of entire inlet is
important to fairness of the study. I create quadrat coverage with
800 ft, this quadrat shows number of points per 64,000 square ft. Based on
this quadrat, I could calculate the spatial randomness of the Loxahatchee Inlet
Arc:
create quad1
Arc: ae
Arcedit:
disp 9999
Arcedit: ec quad1
Arcedit:
backc pollutant 3
Arcedit:
backc point
Arcedit:
mape point
Arcedit:
draw
Arcedit: &r/home/ssa10/gridder
800
Arcedit:
save
Arcedit: q
Arc:
clean quad1
- 800 ft is a empirical number.
Linear regression:
Using linear regression, I found correlation between pollutant
concentration and pollutant sources. Acr/Info doesn't recognize shp
files, but this Loxahatchee data only has shp-file coverage. Therefore,
I
have to be converted into ARC coverages.
Arc:
shapearc pollutants pollutants
Arc:
build pollutants point
Arc:
shapearc summerhouses summerhouses
Arc:
build shouses point
Arc:
shapearc main path main path
Arc:
build main path line
Arc:
addxy pollutants
Arc:
addxy summerhouses
After this procedure, I calculate correlation between concentration
levels and pollutant sources. Therefore, I need to find the distance
between concentration levels and pollutant sources.
Arc:
near pollutants main path line 5000
Arc: near
pollutants summerhouses point 5000
- This will establish the distance from every point in the pollutants
coverage
to the main path and summerhouses coverage and add that distance
to the
attribute table
X - gobi
This program is one Arcview's extension pack. This program produce to
the dynamic changing of symbols or colors in one plot which
simultaneously change corresponding points in other plots or images in
Arcview program. Therefore, I will use this for visualizing pollutants
concentration levels and pollutants sources.
This study conducted with UNIX based Arcview, Arc/Info, and X-gobi
programs, and geoprocessing, spatial analyst, X-gobi Arcview's
extensions are used.
Result
Spatial randomness
Lamda
= 63/70 = 0.875
P(0)
= 0.417 = 30.014
P(1)
= 0.365 = 26.271
P(2+)
= 0.160 = 11.494
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Based on the calculation the chi - square is 4.394 with 2 degree of
freedom. Therefore, I conclude that the sample points are randomly
distribute throughout the inlet with 90% confidence interval
Correlation between pollutants concentration and main boat path
I use a regression coefficient between pollutants concentration and main
path
of the boat traffic.
|
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
Based
on data, I can calculate
r =
{sigma [(concentration - average concentration)*(distance - average
distance)] } / [(N * Standard Deviation of X * Standard Deviation of Y)]
= -
147267.91/63(66.08)(907.421)
=
- 0.38, and r square = 0.14
It means negative relationships between pollutants concentration and
main boat path. The r square is only 14%, which means 14% of the Y variation
is explained by X. Therefore, pollutants concentration and main boat
path
doesnĄ¯t show strong relationship.
Concentration of the pollutants VS
Summerhouse outlet
|
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
Yhat
= a+bXi
b=
-1417797.696/261518832.449 = -0.005
a =
74.17 - (-0.005*2634.124) = 87.341
r =
25824918106/132594386709 = 0.20 and r square = 0.04
This test indicate only 4% variable in pollutant concentration related
to the distance from the summerhouse outlet. This test also doesnĄ¯t
show the relationship between pollutant concentration and summerhouse
outlet
Concentration and Distance from the Boat
Main Path
These pictures are shows pollutant concentration and main boat path result.
Using brush function in X-Gobi, red cross represent sample point, and
green rectangle shows high concentration point. The most of high
concentration area is left-hand side
Concentration and Bottom types
These graphs show relationship pollutant concentration and
bottom types. Numbers (1 to 5) represent each particle size, and 1 is
biggest particle size. This result shows very close relationship
between these two variations.
Concentration and Depth
Pollutant concentration is strongly related to the depth, because 4 is
deepest water in this inlet. The most of higher concentration is
located in the shallow water, so water depth might be key element for
the
diffusion process.
Pollutant Concentration and Distance
from the summerhouse outlet
These images doesn't show much about the relationship between these two
data, but the many of high pollutant concentration is located near by
summerhouse outlet. Furthermore, those two biggest pollutant concentration
sample points are closet location from the summerhouse outlet.
Conclusion
This Loxahatchee Inlet data shows certain relationship between the
diffusion of pollutants, but it's hard to say that surrounding
environment strongly influence to pollutants diffusion processes.
Bottom types and depth result shows very close relationship to the
pollutant concentration. In addition, the most of higher concentration
sample points are located near Loxahatchee river, it's might be the
bottleneck effects. However, I don't have much additional such as
water flow, more sample point, exact pollutant material, and biological
aspect. If I have some more data for this study, I can clearly define
the
process to the Loxahatchee Inlet, or I miss focused this data set.
References:
Wolman, M. G. (1988). Changing national water quality polices,
T. WPCF
60(10): 1774 - 1781
P.E. Rijtema, V. Elias (1996) Regional Approaches to Water
Pollution
in
the Environment, NATO ASI Series
Jay Devore (1999) Applied Statistics for Engineer and
Scientist,
Brooks/Cole Publishing Company
Cook, D., Symanzik, J., Majure J. J., and Cressie N. (1997).
Dynamic
Graphics in a GIS: More Examples Using Linked Software. Computer and
Geosciences: Special issues on Exploratory Cartographic Visualization,
23(4):371 - 385. Paper, CD, and Http:// www.elsevier.nl/locate/cgvis