Agriculture Reference
In-Depth Information
difficult to assign to a particular stream watershed based on surface topography.
Instead, analysis of groundwater levels in domestic water-supply wells can be used
to reveal the direction and magnitude of underground flow and can provide a rea-
sonable indication of watershed boundaries (Lusch 2009).
Hydrologic budgets for the Kalamazoo River watershed (Allen et al. 1972) show
for a 34-year record (1933-1966) that of the period's 889 mm of annual precipita-
tion, about 580 mm (65%) was returned to the atmosphere by evapotranspiration.
Most of the remainder became river runoff, mainly via groundwater flow paths. The
annual rate of groundwater recharge by precipitation in the area averaged 229 mm
(26% of mean annual precipitation) and occurred mainly during the cooler months
of November through May, when evapotranspiration rates are low (Fig. 11.3).
In general, surface runoff is low because glacial deposits in the watershed have
high hydraulic conductivity, which facilitates infiltration. Thus, on a landscape
scale most precipitation that is not evapotranspired readily infiltrates the soil and
percolates to the water table, making groundwater flow paths especially important.
Groundwater flow in turn supplies water to lakes, streams, and rivers. The hydro-
logical linkages between ground and surface waters in the KBS LTER landscape
resemble those well documented in the North Temperate Lakes LTER site in north-
ern Wisconsin (Webster et al. 2006).
A hydrologic budget has been constructed for Augusta Creek (Fig. 11.1), the tribu-
tary of the Kalamazoo River draining ~90 km
2
of land including the eastern side of the
KBS LTER sites (Rheaume 1990). For a year of average precipitation (1977: 950 mm),
evapotranspiration returned 65% of the annual precipitation to the atmosphere (Fig.
11.4), the same percentage found by Allen et al. (1972) in earlier work on the entire
Kalamazoo River watershed. The remaining 35% was discharged as stream runoff,
of which an estimated 75% entered as groundwater flow, corresponding to an annual
groundwater recharge rate of 248 mm (26% of precipitation in that year). Thus, the
area around KBS, which includes the Main Cropping System Experiment (MCSE) of
the KBS LTER (Robertson and Hamilton 2015, Chapter 1 in this volume), has a simi-
lar groundwater recharge rate to that of the overall landscape of the region.
While such hydrologic budgets provide a long-term water balance, they give
no picture of the time scales of water movement through landscapes (Webster
et al. 2006). Given the importance of groundwater flow paths and the large vol-
ume of groundwater reservoirs, transit times of water through these watersheds
are undoubtedly long compared to watersheds in which overland flow is a more
important route for water movement. Groundwater dating—using tracers such as
industrial chlorofluorocarbon gases—shows that the mean age of groundwater
sampled from water-supply wells in recently glaciated landscapes is often several
decades (Saad 2008, Rupert 2008, Stewart et al. 2010, Hamilton 2012). Although
the mean age of groundwater discharged into streams can be similar to that of
water from groundwater wells, streams typically receive convergent groundwa-
ter flow paths of widely differing ages (Böhlke 2002). While groundwater dat-
ing based on tracers has not been conducted in the vicinity of KBS, a model
for the Augusta Creek watershed shows the expected convergence of flow paths
of widely varying distances and residence times to deliver water to the stream
(Bartholic et al. 2007).
