Measuring Water Resources Unit 2.1: Student Exercise – High Plains AquiferJon Harvey (Fort Lewis College) and Becca Walker (Mt. San Antonio College)

Now that you are well-versed in the water cycle and the myriad reservoirs and transport pathways it comprises, it is time to look at some real data from real locations with real users and real implications when water supplies are compromised.

IntroductionLet’s start by thinking about groundwater. Groundwater is stored in aquifers, which can be as small as the ribbon of sand and gravel under a creekbed to vast rock formations hundreds of feet thick and hundreds of miles wide. Water is stored in the microscopic pore space of the rock, which, although difficult to see, can store enormous amounts of water beneath our feet.

Aquifers are recharged primarily through precipitation that infiltrates the soil and enters the pore space in the subsurface materials. Additionally, water from streams and rivers can recharge aquifers by sinking into the streambed (more common in drier climates).

Water users around the world rely on aquifers to provide reliable, clean, and abundant water. They access the water either by collecting it from springs (which occur where the aquifer intersects the ground surface) or by drilling wells into the aquifer and pumping the water to the surface. Many important aquifers around the world are facing increasing demand - can the supply keep up with it? For example, in many aquifers in the U.S., demand has outpaced aquifer recharge, causing aquifer depletion that threatens the sustainability of groundwater as a reliable water resource (Figure 1).

We can keep track of how much water is in an aquifer (its water storage), with this equation:

ΔS = P - ET - R - W

Change in water Storage = Precipitation - EvapoTranspiration - Loss to Rivers – Withdrawals

Although the equation is relatively simple, we can only calculate changes to storage if we have accurate estimates for the terms on the right side of the equation:

Questions or comments please contact education AT unavco.org. Version Dec 30, 2018. Page 1

Figure 1. Cumulative loss of water from major aquifer systems or groups in the United States, 1900-2008. (From Konikow, 2011, Geophysical Research Letters).

Unit 2.1: Student Exercise – High Plains Aquifer

Precipitation - We can measure this with rain gauges and Snotel stations, but it is rather spatially variable, and difficult to accurately capture over a whole watershed.

Evapotranspiration - We can model this using algorithms based on land cover, vegetation type, soil moisture, and weather/climatic conditions experienced at the site, however, there is substantial uncertainty in model predictions, limited by sparse data inputs.

Rivers - The loss/gain of groundwater from/to rivers can be estimated by comparing river discharge measurements in the downstream direction. Only possible on gaged streams.

Withdrawals - Hard to keep track of unless users are voluntarily reporting their withdrawals

With so many unknowns in the equation, changes in the total volume of water in the aquifer can be difficult to predict. That is where some of the measurement techniques you learned about in Unit 1 can fill in the story. These methods attempt to more directly measure the left side of the equation: the underground water storage itself. These methods include:

-Depth to water table in water wells-GRACE satellite observations-Vertical GPS measurements

Questions or comments please contact education AT unavco.org. Version Dec 30, 2018. Page 2

Unit 2.1: Student Exercise – High Plains Aquifer

REVIEWBelow are brief synopses of the groundwater measurement techniques that you will be using in this exercise. For more details, see the materials from Unit 1 and the accompanying powerpoint slideshow.

Measuring changes in water storage with depth to water table in wells The traditional way to track changes in water storage is by directly measuring just how

deep the water table is beneath the surface. When the depth to water table decreases (i.e. groundwater table is rising in elevation), the aquifer is refilling. When the depth to water table increases (i.e. groundwater table elevation is falling), the aquifer is losing water. One challenge of this data source is that water table elevations are very sensitive to local withdrawals and recharge, and aquifer complexity may mean that a well in one location could behave very differently than one only a few miles away. In order to track regional water storage changes, one must include data from several wells across a region.

Measuring changes in water storage with GRACE satellite The GRACE satellite is equipped to measure subtle changes in the strength of gravity

across Earth’s surface (spatial patterns) and over time (temporal patterns). Although many factors influence just how hard the tug of gravity is across the surface, one factor exhibiting temporal changes is how much water is stored in the aquifers beneath our feet at a given time. Measurements from GRACE help scientists estimate the change in water storage in large aquifers, including the HPA.

GRACE data is commonly processed to show changing amounts of water storage, in units of Water Equivalent Height (WEH). This is the equivalent depth of water that is extra / missing from the crust at that location. For example, A WEH anomaly of -10 cm means that the amount of water missing from the crust at that location is equivalent to removing 10 cm of standing water from everywhere in that location.

Measuring changes in water storage with vertical GPS Vertical GPS stations tell us about tiny vertical movements of the ground surface. How

does that help us with groundwater? In many bedrock aquifers, including the HPA, as water storage increases, the ground surface should sink slightly under all that weight. Conversely, we would expect the surface to rebound slightly as aquifer water storage decreases (need to be careful, though - clay-rich aquifers expand when they gain water and contract as they are pumped out, leading to the opposite effect). Scientists track this vertical motion using networks of permanent GPS stations like the Plate Boundary Observatory. Similar to the depth-to-groundwater data, GPS data is station-based. To study changes in groundwater storage across broad areas, several different GPS stations must be included in the analysis.

Together, these tools help scientists and water managers keep account of water resources and how they are changing through time. Long-term studies of water resources in major aquifers in the US show that many are losing large volumes of water (Figure 1). Among the most depleted is the High Plains Aquifer.

Questions or comments please contact education AT unavco.org. Version Dec 30, 2018. Page 3

Unit 2.1: Student Exercise – High Plains Aquifer

INTRODUCTION TO THE HIGH PLAINS AQUIFERIn this unit, you will explore one of the largest and most important aquifers in the United States - the High Plains Aquifer (HPA). Underlying portions of eight states in the central US, the HPA is a shallowly-buried apron of gravel and sand that that was shed off the growing Rocky Mountains (Figure 2). Its water is used to irrigate millions of acres of grain in America’s “Bread Basket” – In fact, nearly a third of the United States’ irrigated crops rely on the HPA. That grain goes on to be processed into food for humans, food for livestock, and ethanol production. In addition, over 80 percent of the residents within the HPA rely on it for their domestic drinking water. Clearly, the HPA is of great importance not only to the residents of the High Plains but for the whole US economy!

Users of HPA water primarily access it by pumping from groundwater wells drilled 10s to 100s of feet into the HPA. Unfortunately, across much of the HPA pumping withdrawals have exceeded natural recharge of the aquifer, so the aquifer is being depleted (water levels are dropping) (Figures 1, 4). Some studies show that much of the water in the HPA entered it during the last Ice Age and can thus be considered ‘fossil water’ (i.e. not being renewed).

Read more about changes in HPA water storage from a USGS report here (don’t worry, it’s only one page long!): https://ne.water.usgs.gov/ogw/hpwlms/files/HPAq_WLC_pd_2013_SIR_2014_5218_pubs_brief.pdf

Questions or comments please contact education AT unavco.org. Version Dec 30, 2018. Page 4

Figure 2. Map showing location of High Plains Aquifer in gray. Locations of groundwater stations (blue dots) and GPS station (red dot) used in the assignment are included.

Unit 2.1: Student Exercise – High Plains Aquifer

1. By how much has the water table declined (1950 to 2013) in the most depleted parts of the HPA? (hint: look at the map in the above link)

2. What was the volume of water lost from the HPA during that time (pre-development to 2013) measured in acre-feet? (hint: read the bullet-point summary at the top of the report.)

3. Your answer to #2 is probably a difficult number to digest…let’s try to visualize it by imagining that we made a swimming pool the size of the state of Kansas (area = ~52 million acres). If we could put the volume of water in your previous answer into that pool, how deep would that water be? (hint: volume = area * depth)

Hopefully that has impressed upon you that we have already pumped an enormous amount of that ‘fossil’ water from the HPA. So water managers are now faced with how to manage a dwindling resource.

Questions or comments please contact education AT unavco.org. Version Dec 30, 2018. Page 5

Unit 2.1: Student Exercise – High Plains Aquifer

OPTIONAL Read this article about challenges facing the HPA, and potential solutions: https://www.geosociety.org/gsatoday/archive/27/6/pdf/GSATG318GW.1.pdf

4. List three ways that HPA states are trying to curb losses in HPA water.

Unit 2.1 – Measuring Water Resources in the High Plains Aquifer (HPA)In the spreadsheet provided by your instructor, you will find historic hydrologic data showing how water storage in the High Plains Aquifer has changed as seen from the multiple methods of measuring groundwater storage described above (page 2). Open up the spreadsheet file and take a look!

Questions or comments please contact education AT unavco.org. Version Dec 30, 2018. Page 6

Unit 2.1: Student Exercise – High Plains Aquifer

You should see several tabs at the base of the document - each takes you to a different spreadsheet in this Excel Workbook.

This first tab, GW-1, shows depth-to-water-table (DTWT) data from a well in southwest Nebraska. This well has data going back to 1986. You will also see a chart with a blue line running across it. This plot is showing the actual depth (in feet) to the water table through time, and shows data from 2000 to 2018.

5. Without any specialized technology, how one could measure DTWT in a simple backyard water well? Include a sketch in your answer.

Depth-to-water-table sensors on wells are one component of the United States Geological Survey’s (USGS) Water Data program.

Here is a map showing the USGS’ observation network for the HPA:https://groundwaterwatch.usgs.gov/net/OGWNetwork.asp?ncd=hpn

Try this link to see where else we have similar water data in the rest of the US:https://groundwaterwatch.usgs.gov/default.asp

Now let’s get back into the data in sheet GW-1. This station in central Nebraska (See Figure 2 for location) exhibits some interesting behavior that we can use to learn about how the annual water cycle works in this location. Examine the X-axis to determine where a given calendar year starts and ends. Now look at the blue line showing the position of the water table over time. Note that if you ‘mouse-over’ the curve you should see a pop-up with the date and value for the closest point (ask your instructor how to view the data points if not using Excel).

Questions or comments please contact education AT unavco.org. Version Dec 30, 2018. Page 7

Figure 3. Screenshot showing tabs at base of workbook.

Unit 2.1: Student Exercise – High Plains Aquifer

6. Can you make out any annual and/or seasonal cycles in the depth to the water table? These might look like movement in a similar direction at a similar time in most years. If so, describe them and speculate on what might be controlling them (hint – look at Figure 4 to right)?

7. List any years that look like they could have been bad drought years:

Questions or comments please contact education AT unavco.org. Version Dec 30, 2018. Page 8

Figure 4. Climate chart for the area near GW-1

Unit 2.1: Student Exercise – High Plains Aquifer

One common way to keep track of wet and dry years is with the USA Drought Monitor. Published weekly, this drought assessment map shows how drought conditions vary across the country. An example map for the conterminous US from December 2018 is shown below in Figure 5. Notice that darker shades of orange/red indicate more extreme drought conditions in that area:

Now look at Figure 6, which shows the % of the HPA in a drought state at a given time:

Questions or comments please contact education AT unavco.org. Version Dec 30, 2018. Page 9

Figure 6. Chart showing percent of HPA area that is in a state of drought between 2000 and 2019. Periods with a lot of orange-red-maroon colors are times of widespread severe drought. (use Word to zoom in if the axes are illegible)

Figure 5. Map showing drought areas in the conterminous U.S. as of Dec 25, 2018

Unit 2.1: Student Exercise – High Plains Aquifer

8. Based on figure 6, list any periods of time that look particularly dry in the HPA:

9. Now look back to your chart for GW-1. How did the groundwater table respond in those years of more significant drought?

10. Based on your answers to the above questions based on well GW-1, what can you say about the long-term health of the aquifer at this location?

Questions or comments please contact education AT unavco.org. Version Dec 30, 2018. Page 10

Unit 2.1: Student Exercise – High Plains Aquifer

OK - Now let’s take a look at another nearby groundwater monitoring well. Data from this well is displayed on tab GW-2. This well is in NW Kansas, still within the HPA (See Figure 2 for location).

11. Can you discern any short-term (annual and/or seasonal) cycles in groundwater levels? If so, how do they compare to what you saw in GW-1 in terms of timing within the year?

12. Describe any long-term (multi-year) trends in groundwater levels. How does that compare to what you saw in GW-1?

What could explain the differences between GW-1 and GW-2 if they are both in the HPA? Often such big differences are the result of the wells tapping water stored in different sub-aquifers that may receive water recharge from different sources and over different timelines.

13. Look at the map in Figure 1, and the websites for the two stations (GW1: Click Here and GW2: Click Here). Compare the locations of the wells, how deep they were drilled, what ‘local aquifer’ they were completed in, and whether or not there is a large river nearby.

Well GW-1 GW-2Depth    Local Aquifer    Near River?

Questions or comments please contact education AT unavco.org. Version Dec 30, 2018. Page 11

Unit 2.1: Student Exercise – High Plains Aquifer

14. *Think Deeper Question* Describe the difference in reliability of water derived from GW-1 and GW-2, and explain those differences using what you have learned about those wells, about the water cycle, and about the history/status of the HPA. Take a paragraph or two to explain your answer, referring to specific features in the plots that led you to your conclusion.

Now we can say something about the status of the HPA in the locations sampled by wells GW-1 and GW-2. Although it is great to have hard data from those locations, sometimes we might want to know about groundwater supplies in places where wells are few and far between. That’s where other datasets can come in handy.

Let’s look at how the groundwater data compares with data derived from the GRACE satellite. Remember, the one that measures small changes in gravity due to redistribution of mass around on Earth’s surface? You may recall that one of the exciting applications of GRACE is to use it to estimate changes in water storage over broad areas.

For the sheets GW-1 and GW-2, you can ‘activate’ GRACE data on each plot by:-Right-clicking somewhere in the plot area-Select ‘Select Data’-Check the box for GRACE.-You should now see an orange line on the plot that shows GRACE data as water storage anomaly (difference from normal) of Water Equivalent Height (WEH, shown in cm). Values of WEH shows on the alternate Y-axis to the right.-If you are unable to turn on the GRACE data, ask your instructor for help.

Questions or comments please contact education AT unavco.org. Version Dec 30, 2018. Page 12

Unit 2.1: Student Exercise – High Plains Aquifer

15. Check back to your notes from Unit 1 and the summary on page 3 of this assignment - what is Water Equivalent Height as measured by GRACE? What does it mean for water storage when the measured WEH increases or decreases?

16. How well does the GRACE curve match the DTWT data for GW-1? What about the particular wet and dry years you identified in question 7? Do they show up as you might expect them to in GRACE data? Explain.

OK. Now let’s zoom out and look at long-term water storage of the HPA as a whole. Go to the tab ‘WetDry’. Here is a plot showing an average of GRACE WEH data for the entire HPA. This plot should, theoretically, tell us about long-term changes in water storage in the aquifer as a whole.

17. Describe the typical annual cycle in terms of the seasonal timing of increasing vs decreasing water storage.

Questions or comments please contact education AT unavco.org. Version Dec 30, 2018. Page 13

Unit 2.1: Student Exercise – High Plains Aquifer

You are also presented with two ‘Drought Monitor’ maps - these are produced every week by the USDA to help the public keep track of developing and prolonged drought problems. These dates represent particularly wet (May 2010) and dry (Nov 2012) times in the HPA for our consideration.

18. Describe the drought status of the HPA region at those two dates.

19. Find the two dates (May 2010 and Nov 2012) as colored dots on the GRACE plot. What Water-Equivalent Height values correspond to those dates (mouse over them to get exact values)? Keep in mind that negative values imply water deficit.

So, here we have two dates with starkly different water storage values as measured by GRACE. We can tell that water storage decreased between May 2010 and Oct 2012. That’s good, but to put it into context, let’s figure out the actual volume of water that was lost from the HPA region during that time. In order to do that we will have to multiply the difference in WEH (depth) between those dates by the area of the HPA (~453,000 km2).

20. Calculate the volume of water lost between May 2010 and Oct 2012 in cubic kilometers (km3). (hint – you must first convert the WEH to km by dividing by 1,000,000, the number of mm in one km).

Questions or comments please contact education AT unavco.org. Version Dec 30, 2018. Page 14

Unit 2.1: Student Exercise – High Plains Aquifer

21. Let’s compare that number to some other volumes of water that may help with understanding just how much water was lost during that period.

a. According to the EPA, the average American family uses around 100,000 gallons per year. For a city of 1 million families, that equates to ~0.38 km3/yr. For how many years could a city of that size operate if they had access to the volume of water lost from the HPA from 2010 to 2012?

b. A typical Olympic swimming pool has a volume of ~0.0000025 km3. How many Olympic swimming pools were lost from the HPA during the period in question?

OK - Let’s look at the HPA from one final perspective - that of VERTICAL GPS. This system helps scientists track movements of the ground surface over time. Vertical GPS can be used to track water resources by detecting tiny vertical shifts in the ground surface as a result of changing water storage. This works because the weight of groundwater tends causes the ground to sink ever so slightly, and when that weight is removed the ground springs back up. Although that relationship holds true in many regional aquifers, it should be noted that in some aquifers, like in the Central Valley of California, we get the opposite response due to the presence of compactible clay in the aquifer which contracts as water is pumped out, causing the ground above to sink.

Take a look at the map of GPS stations within UNAVCO’s Plate Boundary Observatory at this page: https://www.unavco.org/instrumentation/networks/status/pbo

Questions or comments please contact education AT unavco.org. Version Dec 30, 2018. Page 15

Unit 2.1: Student Exercise – High Plains Aquifer

22. Explore the map of GPS stations. Where do we have the best coverage? Worst? How is the coverage in the HPA?

Now let’s look at the data from station P039 - which is situated in far-northeastern New Mexico (Figure 2). These data can be found on the tab ‘GPS’ in the Excel file. When the Y-axis value goes up, the GPS station rose vertically by that amount.

23. What does it likely mean for water storage when the Y-axis value goes up? (hint – the HPA aquifer is generally not a clay-rich, compactible aquifer).

24. Can you make out any long-term trends? Explain how water storage in the area is changing according to the vertical GPS data.

Questions or comments please contact education AT unavco.org. Version Dec 30, 2018. Page 16

Unit 2.1: Student Exercise – High Plains Aquifer

Right-click the plot and add the GRACE data as you have in the other plots.

25. Do the GRACE data support your answer to question 24? If not, explain.

OK, now think back to all that you have covered in this unit so far. Summing it up:

26. In this unit you looked at three methods for keeping tabs on underground water storage.a. Which method(s) is/are most appropriate for keeping track of local groundwater

supplies?

b. Which method(s) is/are more appropriate for keeping track of long-term trends in regional aquifers like the HPA?

27. How could farmers who rely on the HPA for irrigation prepare for predicted increases in water scarcity in the future? (there are many possible answers for this one)

Questions or comments please contact education AT unavco.org. Version Dec 30, 2018. Page 17