Background
San Luis Valley (SLV), in southern Colorado, is
predominantly an agricultural community. Throughout the years, the USGS has
been interested in the effects of the continuous pumping of water from the very
shallow unconfined aquifer located in the valley. In 1993, with funding from
the National Water Quality Assessment program the USGS drilled 34 monitoring
wells throughout the valley, with the majority of the wells drilled in the
northern part of the valley. GIS was used to pick random locations, with zero
bias, in the northern part of the valley. Water samples were collected from
these wells and were tested in close to 300 different fields. Major sampling
events took place in 1993, 2000 and 2007.
Process
I was able to get all the data collected for the SLV
project throughout the years from an USGS database called ADAPS. I was also
able to get 4 DEM files from our internal GIS map library. These 4 DEM’s when mosaicked
together covered the entire study site. The cover page has the mosaicked DEM’s,
with a color ramp symbology applied to the elevations, placed over the
satellite imagery basemap that comes with GIS. I purposely zoomed out far
enough so that parts of the map did not have the DEM overlay. This was done to
show how a simple DEM with the appropriate color ramp symbology can really
emphasize map features, such as mountains and valley. The coordinate system
that I ended up using was the NAD_1983_UTM_Zone_13N system with the central
meridian set to -105.
I thought I would have an easier time starting this
project because I was able to get all my data with relative ease, but this was
not the case. I pretty much had data overload! I had to spend a fair amount of
time whittling down my data set into something I could use in GIS. I ended up having
to break down each data set year (1993, 2000 and 2007) into separate excel
spread sheets. This was after I selected certain data fields that I felt would
tell the best story on a map.
Once I got my data imported into GIS I turned them into
shape files. I then started to look into what type of spatial interpolation
tool I wanted to use. I ended up using the Natural Neighbor tool because it
only uses a subset of samples that surround a point. This form of interpolating,
to quote the GIS help file, “does not infer trends and will not produce peaks,
pits, ridges, or valleys that are not already represented by the input
samples.” I was able to use this tool mainly because of my large data set over
a relatively small area.
After I applied this tool to each of the selected fields
for all three years I then had to adjust the range and increments scale for
each field so that all three years had identical ranges and increments. With
each year having the identical scale it would be far easier to discern the
temporal differences.
Results
Figure
1: Depth to Water Level change through the years
As seen in Figure 1 the water level has stayed the same
in some areas and dropped as much as 20 feet in others, with the greatest drops
found in the northeast part of the valley, which is dominated by agriculture.
Figure
2: Changes in Alkalinity measured as mg/l as CaCO3
The idea range for alkalinity for drinking water is
60-120 mg/l. When alkalinity is greater than 150 mg/l lime buildup in pipes
will start to form and when alkalinity less than 60 mg/l the water will start
to corrode non-pvc pipes. As seen in Figure 2 the majority of the western part
valley falls into the acceptable range, especially in later dates.
Figure
3: Specific Conductance change through the years
Specific conductance (SC) is the ability of water to
conduct an electrical charge. The more dissolved solids present in the water
the higher the SC levels will be. As seen in Figure 3 the SC levels were increasing
from 1993 to 2000 and then decreased by roughly 2000 
/cm (micro Siemens per cm) in 2007. This is a rather substantial
decrease in the amount of dissolved solids. All of the following figures are of
different dissolved solids.
Figure
4: Sulfate concentration change though the years
The sulfate contours almost mirror the SC contours
through the years. Sulfate concentration levels in the southern part of the
valley were at extremely high levels in 1993 and 2000. The EPA has a Secondary
Maximum Contaminant Level (SMCL) of 250 mg/l for sulfate. SMCL are not enforced
but are recommendations, typically for aesthetic reasons. Although it should be
noted that levels greater than 500 mg/l are unhealthy for infants.
Figure
5: Magnesium concentration change though the years
Figure
6: Calcium concentration change though the years
The concentration levels of Magnesium and Calcium in 1993
and 2000 wouldn’t lead to any negative health effects, but the water would be
rather hard.
Figure
7: Sodium concentration change though the years
Sodium concentration levels have also greatly decreased
since 2000, but not as much as some of the other dissolved solids. EPA
recommends 20 mg/l for a low sodium diet. The 2000 concentration levels,
especially in the southern part of the valley and the one small plume in the
middle north of the valley, are significantly higher than the recommended
levels.
Figure
8: Chloride concentration change though the years
The Chloride concentration levels have decreased through
the years, but not as great as some of the other dissolved solids. The location
of the high concentration levels has also shifted more towards the middle of
the valley through the years. It should be noted that typical natural levels of
chloride are 0-10 mg/l and that anything greater is influenced by humans. This
influence generally caused by septic waste, industrial waste or in this case
probably fertilizer use.
Figure
9: Nitrate concentration change though the years
Nitrate is the only (of what I was able to discern)
dissolved solid that doesn’t follow the disappearance pattern of the other
dissolved solids. The very noticeable (Figure 9) nitrate plume in the middle of
the northern part of the map has concentration levels higher than the EPA’s Max
Contaminant Levels (MCL) for 1993 and 2000. The EPA’s MCL for nitrate is 45
mg/l. Concentration levels in 1993 and 2000 are high enough to possibly cause
Blue Baby Syndrome. High levels of nitrate in groundwater in agricultural areas
can be linked to the amount and type of fertilizer used. In SLV the most
prominent area for farms is also where the nitrate levels are the highest,
although they have decreased since 2000.
Conclusion
After analyzing each field spatially and temporally to
discern patterns or trends I ended up only having more questions, such as:
·
What caused the
large dissolved solids plume in the southern part of the valley to pretty much
disappear from 2000 to 2007?
·
What caused the
high concentration plume of nitrate in the northern part of the valley to
decrease from 2000 to 2007? Even though this is a good thing.
·
Has the decrease
in depth to groundwater throughout most the valley caused any growing problems
for the local farmers
·
Since all the
samples were taken in the heart of the growing season (July/Aug), what would
the water samples portray if they were taken in the middle of the dormant
season (Dec/Jan)?
·
Did the drop in
water level cause the dissolved solids concentration to decrease?
Future Work
My supervisor, Dianna Crilley, is planning on presenting
a slide show to the farmers of SLV this summer with hopes of getting funding in
order to do a winter ground water sampling run. She’s going to use a few of my
images to help convince the local farmers that winter samples would help to
complete the ground water picture, since the farmers, rightly so, are very
concerned with their water supply.