SOURCES OF WATER TO DEEP WELLS IN SOUTHEASTERN WISCONSIN

The ground-water flow model for southeastern Wisconsin was designed to be a tool that quantifies the changing sources of water to wells over time. The most interesting results are for recent conditions, corresponding to year 2000. Here is how the model accounts for the ground water currently discharging to shallow and deep wells within southeastern Wisconsin:

Water transferred from surface water within the 7 counties of southeastern Wisconsin (the SEWRPC region) accounts for 80% of combined shallow and deep pumping.

Most of this water is ground water captured by wells that would otherwise discharge to inland surface water bodies (accounting for 59% of total combined pumping), while another part is water induced by wells into the ground-water system directly from streams and lakes into the ground-water system (12%). Some originates as water that in the absence of wells would discharge directly to Lake Michigan or to the rocks below it (7%) while a small part is drawn directly from the Lake (about 2%). Most of this last component is flowing downward from the Lake toward the deep part of the flow system.

The sources of water for shallow and deep wells are release of ground water from storage below the seven-county region and below Lake Michigan (11%) and ground-water flow into the region (9%). Flow into the region is mostly from the west.

Model output: Sources of water to shallow wells in year 2000	Model output: Sources of water to deep wells in year 2000
(source: D.T. Feinstein, U.S. Geological Survey and K. Bradbury, Wisconsin Geological and Natural History Survey)

One qualification - some of the water transferred from inland surface-water bodies to wells is ground water that under natural conditions would discharge to streams that become part of the river flow that empties into Lake Michigan. This reduction of "indirect" discharge must also be counted as a loss to the Lake.

DIVERSION OF SURFACE WATER - due to pumping

Major and secondary surface-water bodies are both important as sources of water to wells in the seven counties that make up southeastern Wisconsin:

Thumbnail map of major surface-water bodies input to the SEWRPC region of model (large file)	Thumbnail map of minor surface-water bodies input to the SEWRPC region of model (large file)
(source: D.T. Feinstein, U.S. Geological Survey)

Focusing just on the relation between pumping discharge and discharge to surface-water bodies (major, minor, Lake Michigan), we can graph the changes from 1864 to 1950 to 2000:

Model output: Graph of change in amount of discharge by type of sink (source: D.T. Feinstein, U.S. Geological Survey)

DEEP WATER BALANCE - changes due to pumping

In order to understand how source water is routed to deep wells pumping from the sandstone aquifer, it is useful to map the fluxes in and out of the deep part of the flow system. A flux is the volume of water that passes across a boundary in a given amount of time. We use flux units of millions of gallons per day. Before pumping began, the important fluxes included:

1 - Lateral flow across the boundaries for the deep sandstone aquifer below the seven-county southeastern Wisconsin area - in or out
2 - Vertical transfer of water between the shallow and deep parts of the flow system - up or down

Here is a map of the fluxes across the seven-county area that are simulated by the model for 1864 under natural predevelopment conditions:

Model output: Map of predevelopment fluxes in millions of gallons per day to/from deep part of flow system below 7 counties of southeastern Wisconsin (source: Wisconsin Geological and Natural History Survey Open-File Report 2004-01)

nb: The fluxes corresponding to lateral flow occur across vertical sections that penetrates the thickness of the deep part of the flow system. In particular, the lateral flow across the eastern boundary passes through a vertical section through the deep part of the system that follows the curve of the Lake Michigan coastline for Ozaukee, Milwaukee , Racine and Kenosha Counties . nb: The net vertical transfer over the southeastern Wisconsin area taken as whole is downward, that is it leaks from the shallow to the deep part of the flow system. The Maquoketa shale inhibits leakage. For this reason the leakage flux is calculated separately for two areas: where the shale is present and where it is absent.

It is worth remarking that the downward leakage amounted to about 1 million gallon per day over the large area of southeastern Wisconsin where the Maquoketa is present, but amounted to about 4 million gallons per day in the smaller area where it is absent. Also, flux moved out of the deep sandstone aquifer toward the deep rocks below Lake Michigan at a rate of about 3 million gallons per day.

With pumping, two more fluxes appear:

1. Water removed by pumping from deep part of flow system
2. Water removed from resident water within deep part of flow system - that is, water removed from "storage"

Model output: Map of year 2000 fluxes in millions of gallons per day to/from deep part of flow system below 7 counties of southeastern Wisconsin (source: Wisconsin Geological and Natural History Survey Open-File Report 2004-01)

Comparing the two figures, it is evident that:

Downward leakage from shallow to deep rocks has responded to pumping by increasing from about 1 to 9 million gallons per day where the Maquoketa is present and from about 4 to 16 million gallons per day where it is absent. This increase is at the expense of ground-water discharge to surface-water bodies that are connected to the shallow part of the flow system.

The change in the deep flow pattern relative to the Lake Michigan coastline is also dramatic. Where before about 2.8 million gallons per day moved to the east, by 2000 about 3.7 million gallons per day move to the west from beneath the Lake toward inland pumping centers. Most of the water moving from beneath Lake Michigan originates as storage release from the deep rocks below Lake Michigan. It is also interesting to note that despite the increase in pumping, ground water continues under current conditions to leave the deep part of the flow system by lateral flow across its southern boundary. This outward flux is caused by pumping in northeastern Illinois.

Source areas to deep Waukesha wells

The pumping centers withdrawing the most ground water in southeastern Wisconsin are in Waukesha County. Most of these wells are located below the Maquoketa shale in the eastern part of the county. In the western part where the Maquoketa is missing, it is easier for ground water to circulate from the water table to the deep sandstone aquifer. More water is also available because recharge tends to be higher in the western part of the county. The presence of long north-south trending bedrock valleys, where the upper bedrock is eroded and loose sediment is in closer proximity to the deep sandstone aquifer, also facilitates downward leakage. This combination of factors has made for a long-distance relation between the location of many municipal wells in the eastern part of the county and the sources of water to these wells in the western part. The east-west travel paths are typically on the order of 10 miles. The source areas that provide the most water are not necessarily closest to the wells, but instead are associated with high recharge areas near Jefferson County and bedrock valleys that extend north into Washington County.

Model Output: Source areas for deep Waukesha wells. Each block corresponds to one square mile. Block color corresponds to amount of water that circulates from the water table to deep wells (source: D.T. Feinstein, U.S. Geological Survey)

None of the source areas for deep Waukesha wells are located east of the County. This finding presents a paradox because the model shows that for 2000 conditions about one-fifth of the water flowing toward deep wells in Waukesha County enters from the east (on the order of 4.7 million gallons per day). If such a large amount of ground water is entering from the east, why aren't their source areas for the wells associated with Lake Michigan? The answer is that the flow directions converging on pumping centers are very recent relative to the times of travel from the source areas to the wells. The water that is now flowing east into Waukesha County originated far to the west, flowed toward Lake Michigan under the natural gradients that existed before pumping began, and ultimately reached locations east of Waukesha County. Only in the last 100 years did these flow lines reverse direction and curl back across the Waukesha County boundary toward the deep wells in the easternmost part of the County:

Model output: Flow lines for deep ground water that in Year 2000 entered Waukesha County from the Lake Michigan side (source: D.T. Feinstein, U.S. Geological Survey)

In this sense Lake Michigan and the rocks below it are not sources of water for Waukesha wells. The water discharging to the wells all originated as recharge to the water table within Waukesha and surrounding counties. Only in the (distant) future will the deep flow lines that enter Waukesha County from the east actually represent recharge, storage release or induced Lake water originating east of the County. It is true that less water is flowing under and into Lake Michigan than before, but it is important to keep in mind that Lake Michigan water is not actually being pumped from deep wells in southeastern Wisconsin.

Schematically, the current flow regime can be represented like this:

Schematic section showing flow lines to deep wells (source: D.T. Feinstein, U.S. Geological Survey)

DOWNWARD LEAKAGE FROM SHALLOW TO DEEP ROCKS - changes due to pumping

Another way to look at the source of water to wells is to consider the spatial pattern of downward leakage from the shallow part of the flow system to the deep sandstone aquifer.

Color-coded maps show the changes through time for southeastern Wisconsin,
where:
- white indicates no downward leakage or transfer upward from the deep to shallow parts of the system
- blue and green indicate relatively low rates of downward leakage
- yellow and red indicate relatively high rates of downward leakage:

Model output: Thumbnail maps of downward flux from shallow to deep parts of flow system for: A) Predevelopment (no pumping)	Model output: Thumbnail maps of downward flux from shallow to deep parts of flow system for: B) 1950	Model output: Thumbnail maps of downward flux from shallow to deep parts of flow system for: C) 2000
(source: Wisconsin Geological and Natural History Survey Open-File Report 2004-01)

Under predevelopment conditions, ground water moved upward from the deep sandstone aquifer toward the lake over a large area. By 1950 the flow was everywhere downward. The flux of downward leakage is greatest where the Maquoketa shale is absent, such as in western Waukesha County and northern Walworth Counties. There is low rate of downward leakage from shallow to deep rocks at the Lake Michigan coastline. This next graphic summarizes the change in the leakage pattern over the seven-county SEWRPC area. Notice the extent of the Maquoketa shale (dark-blue layer):

Model output: Block diagram of change in downward leakage as a percent of recharge and of precipitation between 1864 and 2000 in southeastern Wisconsin (source: D.T. Feinstein, U.S. Geological Survey)

STORAGE RELEASE - due to pumping

Diversion of shallow ground-water flow from surface water bodies is the most important source of water for pumping from shallow and deep wells. Most of this diversion takes place locally, but some occurs outside southeastern Wisconsin and moves laterally into the area across the boundaries of the seven-county area. Besides diversion of water from surface discharge, there is a second source of water to wells - that is, release from storage due to the draining of pores in unconfined aquifers and due to the compression of the aquifer and expansion of the water in confined aquifers. The rate of storage release is very important in the vicinity of wells when they first start to pump, but generally diminishes in importance over time relative to captured surface discharge.

According to the model results, how much water have shallow and deep wells removed from storage below southeastern Wisconsin?

After 136 years of pumping (1864-2000), the following VOLUMES have been pumped from below the 7 counties in southeastern Wisconsin in BILLIONS OF GALLONS:

Of this volume, the following amounts have been derived from ground water already stored in the rocks:

Comparing the total pumped to the amount drawn from storage, it is evident that other sources of water must exist besides the water that was already in the rocks below southeastern Wisconsin in 1864. In fact most of water over the last 136 years was not derived from storage, but instead from captured baseflow - that is ground water that under natural conditions would discharge to streams and lakes, including Lake Michigan, but instead is diverted to pumping wells. Another, smaller part was derived from storage in a different place - in the deep rocks under Lake Michigan. However, even if pumping has removed a relatively small amount of the total discharge from storage, the amount released between 1864 and 2000 has, in total, been very large. One way to show this is to suppose pumping had ceased completely and instantaneously in 2000, allowing water levels to recover.

How long would it take during recovery to replace 50% of the water drawn out of storage under the seven-county region?

How long would it take during recovery to replace 90% of the water withdrawn from storage?

To next topic --> Effect of pumping on ground-water flow lines in southeastern Wisconsin

return to top