Undergraduate Research in Sustainable Engineering

1
Mascaro Sustainability Initiative
Undergraduate Research in Sustainable Engineering
Mapping of average groundwater levels across the
United States with a focus on data collection and
analysis Heather Pry, Civil and Environmental
Engineering Faculty Advisor, Xu Liang
2
INTRODUCTION
Purpose
To develop a groundwater table mapping approach
capable of accurately predicting groundwater
levels in areas without measured data.
Point of Focus
The State of New Jersey
Methods
Cokriging Principal Component Analysis Darcys
Law applied to small watershed regions
3
CONNECTION TO SUSTAINABILITY
Groundwater makes up more than 95 of the global,
unfrozen fresh water reserves. -According to the
National Ground Water Association and the
Alicante Declaration
Monitor changes in groundwater levels, such
changes could indicate the potential
for Permanent Loss of/Damage To Local Water
Supplies Subsidence Drying of
Wetlands Determine if groundwater is a viable
and sustainable option for use as a source of
drinking water, irrigation, or water for
industrial purposes.
4
DATA COLLECTION AND ANALYSIS
US Groundwater Wells and Number of Measurements
Wells with At Least 100 Measurements
All Wells
5
DATA COLLECTION AND ANALYSIS
Mean Groundwater Tables of NJs 174 Wells with At
Least 12 Measurements
New Jerseys 1,413 Wells
6
DATA COLLECTION AND ANALYSIS
Variables Considered in Correlation Analysis
DATA Groundwater Table Soil Porosity Soil
Hydraulic Conductivity Precipitation
(PPT) Elevation Leaf Area Index (LAI)
SOURCE USGS STATSGO Data STATSGO Data LDAS, 1/8
Degree USGS DEM, 10m Resolution MODIS, Monthly
Data Groupings for Correlation Analysis Include
DATA Soil Type Aquifer Systems Watersheds
SOURCE NOAA NWS New Jersey DEP New Jersey DEP
7
DATA COLLECTION AND ANALYSIS
All wells displayed have at least 12 measurements
8
DATA COLLECTION AND ANALYSIS
The variable most highly correlated with GWT on a
regular basis was Elevation.
9
COKRIGING METHOD
Cokriging
Multivariate, geostatistical interpolation method
used to create a surface Reduces the need for
expensive and time consuming data
collection Makes use of readily available data
which is highly correlated with the primary
variable.
Cokrigings Use in this Study
GWT Data is Limited Elevation Data is
Exhaustive Elevation Data is Free Generally a
High Correlation between GWT and Elevation
10
COKRIGING METHOD
Watershed Management Area 14, Digital Elevation
Model
Watershed Management Area 14, Ordinary Cokriging
Results
Darker Lower Elevation Lighter Higher
Elevation
Lighter Blue Lower Groundwater Table Lighter
Purple Higher Groundwater Table
11
COKRIGING METHOD
Omitted Wells
Cokriging Wells
WMA 14 1453 km2
12
COKRIGING METHOD
New Jersey Watershed Management Area 14, Ordinary
Cokriging Results
13
COKRIGING METHOD
SHORTCOMINGS OF THE COKRIGING METHOD
Statistical not Physical REQUIRES Large Amount
of Data Small Area Strong Relationship
14
PRINCIPAL COMPONENT ANALYSIS (PCA) METHOD
Principal Component Analysis (PCA)
Create Principal Components using the 6 variables
in an attempt to form a linear relationship
between them.
Variables Considered
DATA Groundwater Table Soil Porosity Soil
Hydraulic Conductivity Precipitation
(PPT) Elevation Leaf Area Index (LAI)
SOURCE USGS STATSGO Data STATSGO Data LDAS, 1/8
Degree USGS DEM, 10m Resolution MODIS, Monthly
15
PRINCIPAL COMPONENT ANALYSIS (PCA) METHOD
Watershed Management Area 14 All Wells, PCA
Results
Variance Explained ()
Variance Explained ()
Principal Component
Principal Component
16
PRINCIPAL COMPONENT ANALYSIS (PCA) METHOD
Watershed Management Area 14 All Wells, PCA
Regression
GWT (m)
Wells
17
PRINCIPAL COMPONENT ANALYSIS (PCA) METHOD
SHORTCOMINGS OF THE PCA METHOD
Does not indicate a strong linear relationship
between the variables Results do not provide any
direction for further analysis with this
method. Would require a re-evaluation of the
variables used
18
DARCYS LAW METHOD
Darcys Law
An attempt is being made to use Darcys Law in
order to map the average groundwater table across
small watersheds by approximating baseflow on a
watershed scale.
Variables Considered
DATA Gage Height and Location Stream
Discharge Digital Elevation Model Watershed
SOURCE USGS USGS USGS Delineated using ArcInfo
19
DARCYS LAW METHOD
Darcys Law
q A K dh/dl
q A K (Ht-ht) / (W/4)
q Baseflow A Area of Groundwater Discharge
(q) K Hydraulic Conductivity Ht Groundwater
Table Height ht Gage Height W/4 Length of
Groundwater Flow
20
q A K (Ht-ht) / (W/4)
DARCYS LAW METHOD
Watershed (W)
GW Well1
GW Well2
Stream Gage (q and ht)
BOX 1
GW Well1 June q ht A Ht W K July q ht A Ht W
K K (Constant) Etc. q ht A Ht W K
21
DARCYS LAW METHOD
Advantages
Draws upon regularly measured data Predicts
Close Estimate of GWT (Intended, not proven)
Disadvantages
Cannot Predict Exact GWT Over Simplified Does
not consider Soil Properties in calculating
parameter K Cannot be used in watersheds without
a well Limited to small watersheds (lt100 km2)
22
CONCLUSIONS
Cokriging with GWT and Elevation provides a
suitable means for initial estimation of
groundwater levels to be used as a basis in a
more complicated model. Principal Component
Analysis does not appear to be a promising method
for determining a linear relationship between GWT
and the variables influencing it. The proposed
Darcys Law method, with its assumptions and
simplifications, may also provide a suitable
basis for making initial estimates of groundwater
levels to be used as a basis in a more
complicated model. This method is not completed
or proven and will require additional testing and
modifications.
23
ACKNOWLEDGEMENTS
Thank you to Dr. Liang and PhD student Yaofeng
Yue for their guidance on this project, Thank
you to the Mascaro Sustainability Initiative for
providing this opportunity, Thank you to Gena
Kovalcik, Dr. Eric Beckman, and Kim
Wisniewski for organizing this program, And
thank you to all the Thursday Luncheon speakers
who provided their insight into sustainability,
library use, and graduate school.
24
REFERENCES
Goovaerts, P., 1997. Geostatistics for Natural
Resources Evaluation. Oxford University Press,
New York. 215-258 Goovaerts, P., 1999.
Geostatistical approaches for incorporating
elevation into the spatial interpolation of
rainfall. Journal of Hydrology 228 (2000)
113-129 Mays, Larry W., 2007. Water Resources
Sustainability. McGraw-Hill Mays, Larry W.,
Todd, David K., 2005 Groundwater Hydrology. 3rd
Ed., Wiley. 86-91 Rencher, A.C., 2002. Methods
of Multivariate Analysis. 2nd Ed, Wiley
Interscience. 380-407