Content area
Climate change is expected to increase the frequency and severity of flooding in the Great Lakes region. In many cities, flood‐control infrastructure is insufficient to protect against future climate conditions. Consequently, there is increasing focus on stormwater storage provided by urban greenspace, such as wetlands and prairies, but the ecohydrological behavior of these ecosystems is not well understood when they are embedded within cities. To improve understanding of hydrological connectivity between urban areas and natural greenspaces, we deployed a sensor network in Gensburg Markham Prairie (GMP), a large intact prairie‐wetland complex in south suburban Chicago. We used the resulting high‐frequency time‐series data to assess surface‐subsurface hydrologic dynamics between upland and low‐lying wetland areas, interactions between the prairie and surrounding environment, and stormwater storage provided by the prairie. Rapid infiltration within the prairie during and after storm events provides subsurface flow that stores considerable water, flattens storm hydrographs, and increases the wetland hydroperiod. Much of the stormwater input to GMP derives from the surrounding cityscape. Consequently, storage within the prairie‐wetland system reduces and slows stormwater discharge to downstream urban communities. For a typical 5‐year 24‐hr storm with 10.9 cm of rain, GMP stores 77,100 m3, 64% greater than the estimated direct rainfall volume onto the prairie, yielding 30,000 m3 of offsite stormwater storage. This improved understanding of ecohydrological dynamics in urban prairies and wetlands informs the design and implementation of green infrastructure to meet growing needs for stormwater management.
Introduction
Prairies and wetlands provide valuable temporary water storage and habitat for fauna including pollinators and both endemic and migratory birds. More than half of global wetlands have experienced widespread degradation and loss due to agriculture, drainage, and urbanization, with even higher rates of loss in inland wetlands relative to coastal wetlands (Davidson, 2014). Historically, the Midwest and Great Lakes regions held considerable prairie and wetland habitat, but these habitats have largely been lost over the last 150 years (Dahl, 1990, 2011; Hu et al., 2017; US Army Corps of Engineers, Rock Island District, n.d.). These biomes are often undervalued and rapidly converted, despite the multitude of ecosysterm services they provide (Zedler & Kercher, 2005). Consequently, there has been a focus on restoring and/or constructing wetlands and prairies to return some degree of natural hydrology and ecology, particularly for purposes of urban stormwater control, water quality protection, and biodiversity of native plants, insects, and birds. Small native prairie and wetland systems persist within the highly developed landscapes of the Midwest and Great Lakes regions, but their ecohydrological behavior is not well understood. These remnant prairies and wetlands are surrounded by the built environment and thus not truly undisturbed native systems, but they are also more complex than constructed or restored systems. Thus, while embedded native prairies and wetlands can serve functionally as green infrastructure, their contribution within the larger urban landscape is uncertain.
Understanding hydrological interactions between developed environments and green infrastructure is particularly important given projected changes in climate. In the Midwest and Great Lakes regions, changing regional climate is expected to result in shorter, warmer winters, with greater winter rainfall and less snow cover (Hayhoe et al., 2010; U.S. Global Change Research Program, 2018; Wuebbles et al., 2021). The 20-year storm for the Midwest is expected to increase in precipitation volume by 5%–25% by 2050 (Wuebbles et al., 2014). In the Great Lakes region there has been a 10% increase in precipitation from 1901 to 2015, with more of this precipitation coming from large rain events (Wuebbles et al., 2021). In addition, less snow has fallen across Illinois overall than would be expected from the historical record (Wuebbles et al., 2021). Furthermore, transitions between wet and dry seasons in the Midwest are now occurring more quickly, subjecting ecosystems to dramatic changes (Ford et al., 2021).
The effects of this climate shift have increased flood damage to personal property and public infrastructure. Between 2007 and 2016, flood insurance claims were paid in 95% of the zip codes in Chicago, with most of these claims made in communities of color (Keenan et al., 2020). The locations of these flood damage claims do not coincide with recognized floodplains within Cook County (Center for Neighborhood Technology, 2014; Chicago Metropolitan Agency for Planning, 2018), indicating that the systems resulting in flooding are more complex than simple topography and are related to land development, stormwater infrastructure capacity and management practices. There is also considerable disparity between areas experiencing flooding damage and areas eligible for the National Flood Insurance Program (NFIP), which is limited to floodplains, reflecting a gap in support for the people most affected by urban flooding (Chicago Metropolitan Agency for Planning, 2017). Furthermore, the timing of flooding events is changing, with rain-on-snow events becoming more common in regions which formerly experienced only frozen precipitation during the winter (Beniston & Stoffel, 2016; Surfleet & Tullos, 2013).
Current flood-control infrastructure is not sufficient to protect against these conditions, as stormwater routing and retention structures were designed for past and current climate. The prospect of increasing frequency and intensity of extreme storms combined with undersized and outdated stormwater infrastructure make flooding the single largest concern for urban sustainability and resilience in the region (Ando et al., 2020; Borden et al., 2007; Davenport et al., 2021; Dunn & Miller, 2020; Hunt & Watkiss, 2011). Modifying conventional stormwater infrastructure to reduce flood vulnerability is challenging and costly. These issues have increased interest in urban green infrastructure (GI) and other types of nature-based solutions (NBS) for stormwater storage and retention (Eaton, 2018; Green et al., 2021; Maragno et al., 2018; Rizzo et al., 2018). Prior studies nationwide have examined the ability of distributed stormwater management concepts to mitigate flooding damage within urban catchments and identified low-impact development (LID) incorporating GI and NBS as a key component of future stormwater management (Larsen et al., 2016).
Effective design of both regional and local GI requires prediction of hydrological dynamics under future climate and land use. Most urban hydrology modeling efforts have focused on basic landscape characteristics such as overall percentage of impervious cover, but recent studies have begun to explore hydrologic dynamics at the parcel scale and to link parcel-scale interventions to community-scale benefits (Golden & Hoghooghi, 2018). Voter and Loheide (2018) established that local interactions between impervious and pervious areas strongly affect hydrologic response, while Bhaskar et al. (2018) demonstrated that the effects of GI on stormwater are localized and decrease with distance from interventions. Natural areas within urban landscapes are expected to exhibit similar behavior. However, the hydrology of urban natural areas remains understudied, and little data are available to assess interactions between urban areas and embedded natural greenspaces.
To meet this need for improved understanding of hydrological connectivity between urban areas and embedded natural greenspaces, we deployed an extensive sensor network in Gensburg Markham Prairie, a large intact prairie-wetland complex in south suburban Chicago, and used the resulting high-frequency time-series data spanning nearly 4 years from July 2016 to April 2020 to assess surface-subsurface hydrologic dynamics between upland and low-lying wetland areas within the prairie and the interaction between the prairie and the surrounding urban environment. The study time period included wet and dry years with annual rainfall in the Chicago region ranging from 91.4 cm (2016, 4.8 cm less than average) to 125.8 cm (2019, third wettest year on record in Chicago) (National Oceanic and Atmospheric Administration, 2022). Finally, we assessed the performance of the prairie in terms of stormwater retention and the associated potential for natural areas to reduce flooding in nearby urban areas. While natural variations in vegetation and precipitation at GMP are expected over time, the overall hydrologic structure of the site is believed to have remained stable, making these results relevant for present day conditions.
Materials and Methods
Study Site
Gensburg Markham Prairie (GMP) is a large (40 ha), intact prairie remnant located in Markham, IL, 65 km south of downtown Chicago (Figure 1). GMP is one of the best-preserved prairies in Illinois and is part of the larger Indian Boundary Prairies complex (IBP), a network of natural and restored high-quality tallgrass prairies jointly owned and managed by Northeastern Illinois University (NEIU) and The Nature Conservancy (TNC). Markham is home to approximately 11,400 residents with 75.2% of residents self-identifying as Black or African American and a 21.0% poverty rate (U.S. Census Bureau, 2020). Precipitation in the area is approximately 1.0 m/year, with 70% of the precipitation occurring between April-October (National Oceanic and Atmospheric Administration, 2022). Winter precipitation occurs primarily as snow, contributing considerable volumes of water to spring runoff as snowmelt. Consistent with regional climate change, this area has experienced high rainfall and low snow cover through the late winter and early spring in recent years, relative to long-term average conditions (Climate Engine, 2021; Huntington et al., 2017; National Operational Hydrologic Remote Sensing Center, 2004).
[IMAGE OMITTED. SEE PDF]
Markham and nearby communities suffer considerable flooding damage. The region lies within the Little Calumet River watershed, which drains into the Calumet-Saganashkee (Cal-Sag) Channel and subsequently to the Mississippi River via the Chicago River. Markham is part of the Calumet Union Drainage Ditch (CUDD) subwatershed. The Little Calumet River watershed has considerable overbank flooding and local drainage-related flooding, prompting major infrastructure projects by the Metropolitan Water Reclamation District of Greater Chicago (MWRD) to mitigate flood damage (Metropolitan Water Reclamation District of Greater Chicago, 2017). This watershed is also the focus of a new regional stormwater credit trading program, StormStore, to increase investment in GI to mitigate flooding (Metropolitan Planning Council, 2020).
IBP is located within the Chicago Lake Plain region of the Northeastern Morainal natural division, characterized by generally low-lying topography with fine, poorly-drained soils and occasional sandy ridges (Illinois Department of Natural Resources n.d.). GMP has a complex internal structure (Figure 1c) with four prominent prairie types: Midwest Dry Mesic Sand Prairie, Lake Plain Wet Prairie, Sedge Meadow, and Ephemeral Marsh (Hanson, 1975). Vegetation differs starkly in each of these areas because of their differences in topography, soils, drainage characteristics, and water levels. The dry mesic habitat occurs on a paleo sand ridge oriented southeast-to-northwest that bisects the prairie, while the meadow and marsh habitats occur in lower-lying areas. The ridge is composed of Watseka fine sand, and the remainder of the prairie contains Selma loam and Hoopeston sandy loam (Hernandez Gonzalez et al., 2019).
GMP has a network of drainage ditches along its boundaries and incomplete internal ditches intended for a residential development project abandoned in the early 20th century (Figure 1c). An approximately 2 m deep ditch, Bel Aire Creek, separates the prairie from residential areas to the north. This channel conveys stormwater through a culvert underneath Interstate 294 to the east, imposing a west-to-east flow direction, and discharges into the Little Calumet River via Dixie Creek. This channel was previously connected with perpendicular ditches in the interior of the prairie, but NEIU and TNC blocked the connection points by breaking drainage tiles and filling with local soil in 1999 to increase retention of water within the prairie. An ephemeral wetland stores water in the northern part of the prairie throughout the wet season (spring and early summer). The southern part of the prairie is connected to the wetland via a long swale. The prairie is adjacent to and intermittently hydrologically connected to Markham Prairie North and Markham Prairie South, which are more recently reclaimed prairies and currently undergoing restoration. These prairies are part of the IBP complex and are also managed by TNC. They are separated from GMP by ditches and/or roadways that can be overtopped during large precipitation events.
GMP soils and plant communities are largely intact because the interior of the prairie was never plowed or otherwise converted, which is important as prairie soils require centuries to develop and generally formed far before the introduction of intensive agriculture in the Midwest (Hole & Nielsen, 1968; Kucharik, 2007; Thorp, 1948). Each of GMP's habitats supports distinct, diverse, and endemic plant species, insects, waterfowl, and grassland birds (Beaubien, 1994). Additional information on the ecology of GMP is provided in the Supporting Information S1.
Instrumentation
In summer 2016, InSitu LevelTroll 400 water level sensors were deployed in ditches and in piezometers throughout the prairie (Figure 1c). Surface water level sensors (WLS) were mounted on posts installed in ditches and in the center of the northern wetland region of the prairie (WLS2). Soil expansion in the wetland required sensors WLS1, WLS2, and WLS3 to be raised 20 cm in August 2017 to prevent contact with soil in wet season. Twelve subsurface level sensors (WLW 1-10, 12, 14) were installed in piezometers at depths of 1 m below the ground surface (WLW11 was damaged and removed in 2017). One subsurface sensor was deployed at a depth of 2 m below the highest-elevation area of the prairie (WLW13). Water level sensors are aligned along three transects, two longitudinally crossing (a) the wetland and (b) the sand ridge into the swale, and one approximately perpendicular to the sand ridge crossing the sand ridge and wetland. Pressure data from LevelTroll 400 sensors was corrected using barometric pressure from an InSitu BaroTroll collocated with WLW8 to obtain water depth above the sensor, and then added to the surveyed sensor elevation to yield water surface elevation in meters above mean sea level (AMSL, based on NAVD 88). Measurements were recorded every 30 min. Two Sentek HydroScan soil moisture profile probes (SMP1, SMP2) were deployed, one in the sand ridge and one at the southwestern edge of the wetland. Each profile probe contains six stacked sensors at 10, 20, 40, 60, 80, and 100 cm below the ground surface. Soil moisture data is reported as volumetric water content (% VWC).
A Waggle sensor node was installed at TNC's Indian Boundary Prairies site office, 1.3 km away from GMP's wetland, to collect atmospheric and precipitation data (Beckman et al., 2016). Waggle nodes' standard measurements include temperature, relative humidity, and barometric pressure. An Onset 0.01-inch tipping-bucket rain gauge with 6-inch orifice was integrated with the node. Complete and current information on the Waggle sensor node and individual sensors can be found at . The rain gauge was not heated, so data from Crete, Indiana airport (located approximately 300 km from GMP) were used to estimate frozen precipitation and fill data gaps.
Soil Characterization and Soil Moisture
Soil cores were obtained during installation of each piezometer. Soil was excavated to a depth of 1.2 m using a 5 cm diameter hand auger and cores were removed in consecutive 30 cm sections. Sub-samples from each core section were characterized manually and visually in the field according to ASTM D2487 (American Society for Testing and Materials, 2020). This information was used to develop boring logs with 1-cm depth resolution, including information on soil color, organic matter, and secondary constituents. These logs show good agreement with the Soil Survey Geographic database (SSURGO) soil classifications for the prairie (Soil Survey Staff et al., 2015). Soil porosity was estimated using Equation 1: Soil Porosity = 1–(Bulk Density ÷ Particle Density). Particle density was assumed to be 2.65 g/cm3 (Amoozegar et al., 2023). The bulk density was estimated based on the average values for each assigned Unified Soil Classification System (USCS) soil group (United States Department of the Army, 2012). Porosity values were calculated using these estimates of bulk density, as direct measurements were not feasible within the study scope. Estimated soil porosities ranged from 0.29 to 0.35. While this approach provides a reasonable approximation for broad-scale storage estimation, it introduces uncertainty in volumetric calculations. Estimated storage volumes may have error margins on the order of ±10%–20%, a typical range when using regional values for estimation of soil storage (Carsel & Parrish, 1988; Rawls et al., 1982; Tafasca et al., 2020).
The remaining soil was sub-sampled and used for chemical analyses, as reported by Hernandez Gonzalez et al., 2019. Collected cores, core sections, and subsections are identified by an International Geo Sample Number (IGSN) and logged into the System for Earth Sample Registration (SESAR). Soil chemistry data are available via .
Hydrologic Time Series Analysis
The sensor network furnished 46 months of hydrologic data collected at 30-min intervals at 24 locations throughout GMP. Python was used to post-process data collected from all sensors to support time-series and statistical analyses. A script was used to filter data from sensors that were dry (based on pressure measurements) or likely encased in ice (based on temperature measurements). Filtered data were filled and smoothed using locally-weighted scatterplot smoothing (LOWESS) (Perktold et al., 2023) using a 5-hr rolling window to remove instrument noise while preserving natural sharp increases caused by intense rain events. Power spectral analysis was used to identify critical frequencies and scaling relationships in water level, precipitation, and soil moisture at GMP. Power spectra were generated using Lomb-Scargle periodograms, which are useful for identifying periodic signals in data with irregularity in sampling, such as gaps of varying length and frequency (VanderPlas, 2018). This approach was used to account for conditions where wells go dry, where sensors are frozen, and where the water level reaches above the ground surface, all of which reflect important types of hydrologic response. Periodograms were developed for the 78-year time series of precipitation data from Midway Airport (located 21 km from GMP), and the 46-month time series from the emplaced sensor network: water level measurements from wells (WLW 1-10) and surface sensors (WLS 1-8), and two stacked soil moisture sensors (SMP1-2).
To assess hydrologic responses of the prairie to storm events, storm response hydrographs were extracted from the time series. Stormwater events were identified when the slope of the water level time series increased significantly above the noise threshold. Putative event detections were cross-checked against rainfall data to ensure this was a hydrologic response to a precipitation event. Responses were analyzed in terms of time from baseline to the hydrograph peak, relaxation time (time between peak and return to baseline), and water levels before and after the event.
Stormwater Storage and Flow Routing
A workflow developed in the ArcGIS Pro (esri) 3D Analyst Toolbox was used to develop a digital elevation model (DEM) of the prairie from LiDAR data collected during the 2022 flyover of Cook County (Cook County Bureau of Technology, 2023). The DEM was further processed to develop a flow routing model of the prairie's surface water based on flow accumulation, calculating the contribution of upslope surfaces to water depth on downslope surfaces. The DEM was edited by manually altering the elevation of known man-made ditches, which were not adequately captured by LiDAR due to their shallow depth and extensive vegetation. Survey data was used to correct the elevation of these ditches, which exist throughout the prairie year-round and were previously mapped by The Nature Conservancy.
GMP's storage capacity for surface water, that is, the maximum amount of precipitation that the prairie can store above the ground before contributing overland flow to nearby areas, was estimated by applying a depression-filling algorithm to the DEM. This approach models the terrain of the site as nested contours and routes water added to the system to the appropriate contour based on relative elevation. This approach allowed surface water to be variably connected in time, which is useful in prairie wetland systems with quite flat topography (Wu & Lane, 2017). Surface water storage of stormwater was obtained from the DEM and storm hydrographs measured from the sensor network. The maximum surface volume storage capacity of GMP is approximately 24,800 m3. However, the actual volume may be much higher as the digital elevation model may not adequately capture the volume of shallow surface depressions and the ephemeral wetland area due to the extensive vegetation yielding errors in LiDAR measurements, as documented for the drainage network.
Changes in measured groundwater levels were converted into stormwater storage volumes by multiplying the change in groundwater level by the soil porosity at that location and a catchment area associated with each piezometer, delineated using Voronoi polygons. Soil porosity was estimated based on soil type measured locally (see Section 2.3) and extrapolated to the surrounding area.
Results
Surface Flow Patterns and Internal Hydrodynamics
The predominant surface flow path in GMP runs from southeast to northwest connecting the swale and wetland to the Bel Aire Creek ditch on the northern periphery of the prairie, as shown in Figure 2. This flow path parallels the sand ridge, which represents a hydrologic divide in the prairie owing to its relatively high elevation. Approximately 63% of GMP drains into Bel Aire Creek via this flow path. In addition, Bel Aire Creek receives considerable stormwater inflow from the developed areas to the north and small amounts of stormwater from areas to the west. The second major surface flow system in GMP occurs west of the sand ridge, traveling south-eastward into Markham Prairie South and then reentering GMP at the southeast end of the swale. This drainage system is peripherally connected to small amounts of suburban development to the south, which can be seen in the aerial image (Figure 2). Overall, the surface flow system in GMP is controlled principally by the sand ridge that bisects the prairie, the wetland-swale system to the east, and connections with the engineered stormwater drainage systems to the north and south of the prairie. Based on measurements of surface water in the swale, this surface flow system is connected from approximately mid-April to mid-December. Surface water in the prairie is generally frozen and therefore immobile from late January to early April, with intermittent thaw periods yielding brief flood events.
[IMAGE OMITTED. SEE PDF]
During winter and spring, measured water levels were typically above the ground surface in the wetland and swale areas, yielding surface water up to 30 cm deep. Based on aerial photos and vegetation surveys, the ephemeral wetland occupies about 10% of GMP, though surface water varies greatly in area throughout the year, from completely dry in September to extending across nearly half the site in April and May (Figure 1d). Time series of surface water levels are presented in Figure 3 for a ditch on the edge of the swale near the southeastern corner of the prairie (WLS6), further northwest and downstream into the swale (WLS7), and toward the northern edge of the prairie at the center of the wetland (WLS2). Surface water levels in the ditch (WLS6) are much flashier than in the wetland (WLS2), as expected because of the high flow network connectivity both within the site and with the surrounding urban area. This behavior is representative of behavior in ditches throughout the site. The ditch responds more rapidly to precipitation and drains more quickly than further downstream in the swale (WLS7). Water levels in the wetland (WLS2) reflect hydrologic signal filtering across the entire prairie, resulting in an overall damped hydrologic signal in the wetland following precipitation events. The water level in the ditches is lower than in the wetland for most of the year, with the lowest absolute water level occurring at the eastern end of Bel Aire Creek (WLS8), which is the furthest downgradient monitoring location. After this point, the Bel Aire Creek channel joins with the stormwater drainage infrastructure for Interstate 294, ultimately conveying runoff to the Little Calumet River.
[IMAGE OMITTED. SEE PDF]
Surface-Subsurface Hydrologic Dynamics
In the time series data of water levels (Figure 4), hydrologic responses to precipitation are generally rapid with sharp increases in water level, and hydrograph recession is also rapid during the spring and summer. The sensors measure both surface and groundwater levels, allowing continuous and consistent monitoring of surface-groundwater fluctuations across the site. The water table frequently rises above the ground level in the wetland during the period of high precipitation in the spring. When the water table is below the ground surface, it responds strongly to high-intensity storm events, generally rising quickly during storms. This is particularly true across the sand ridge and other highly permeable areas of the prairie (e.g., location WLW7). The wetland (WLW2) has slower hydrograph recession than observed in the sand ridge, which is consistent with the expected slower flow through the wetland soil profile (Figure S1 in Supporting Information S1). Soils at this location include layers of dense, high-plasticity clay and silty, low-plasticity clay, with porosities of 30%–35%.
[IMAGE OMITTED. SEE PDF]
The violin plots in Figure 5 show the distribution of groundwater elevation across the 46 months of data collection for five locations on a transect crossing through the sand ridge to the ephemeral wetland. Separate distributions are shown for each season in order to compare water levels between seasons and assess hydrological dynamics within each season. Water levels are extremely different between the winter-spring (blue, green) and the summer (red) seasons in the wetland areas (WLW4, WLW5), but show extensive seasonal overlap in the sand ridge (WLW7, WLW8). Water levels are more variable in the fall (yellow), and the distribution overlaps with the other seasons. In winter and spring, groundwater levels are highly variable in the sand ridge, consistent with the observed rapid response of the sand ridge to precipitation. Additional violin plots for each season are provided in Figure S2 in Supporting Information S1.
[IMAGE OMITTED. SEE PDF]
In the wetland, water levels in the winter and spring are generally above the ground surface, with only a small fraction of the water level distribution below ground level. This reflects two factors: 1. periods of water input in the absence of significant evapotranspirtation (ET), and 2. water draining from the sand ridge into the adjacent swale and wetland. This hydrologic behavior is typical for the region as it reflects seasonal trends of precipitation. The polar jet stream is often located over Illinois in the winter, spring and fall leading to frequent low pressure storm systems and increased precipitation (Illinois State Water Survey, n.d.). Shifts in the polar jet stream combined with increased cloudiness due to Lake Michigan result in decreased precipitation in the summer (Illinois State Water Survey, n.d.). Decreased precipitation and high evapotranspiration rates from extensive vegetation growth draw down water levels across the prairie in the summer, consistent with prior observations from prairies across the central and northern U.S. (Shjeflo, 1968; Wagle et al., 2017). Consequently, the water table in the wetland falls to more than 1 m below the ground surface in much of the prairie by mid-summer (Figure 4). Similarly, groundwater levels in the sand ridge are often below the depth of the sensors (184.6–185.7 m) in the summer and early fall. In the fall, increased precipitation and decreased ET cause water levels to increase, first yielding higher groundwater levels and then restoring surface water in the wetland (yellow distribution in Figure 5 and Figure S2 in Supporting Information S1). These seasonal transitions in water level are crucial for understanding when and where stormwater retention and connectivity occur and determine the wetland habitat time profile.
Soil moisture probes captured the wetting and drying of soils in response to precipitation, internal flow processes, and ET on both seasonal and diurnal timescales (Figure 6). There are three primary modes for soil moisture in the prairie. In mid-summer through mid-fall (approx. July-November), the soil is drying out from the top down owing to the activity of the predominant prairie vegetation such as grasses and forbs (red boxes in Figure 6). In fall and winter, the prairie wetland is wet, but not saturated. In mid-winter through spring the soil moisture increases to approach saturation, and then does not change until the growing season begins and the prairie wetland begins to dry out again. Consequently, wetland soils are saturated at all observed depths from mid-winter through mid-summer. Subsequent summer drying does not reach the deepest wetland soils except in extremely dry years, such as 2017. In comparison, at the sand ridge, which has high permeability and drains very quickly, the soil rarely becomes fully saturated even during wet spring conditions. These contrasts between habitats show how vegetation-soil interactions influence both short- and long-term water retention potential.
[IMAGE OMITTED. SEE PDF]
The wetland hydroperiod (the length of time the wetland contains surface water) increased from 2017 to 2019. In 2017, the hydroperiod was 4.6 months with the wetland being continuously dry from approximately June-November and intermittently dry in January, February, and December. In 2018, the hydroperiod was 7.6 months and the wetland was dry from August-November. In 2019, the hydroperiod was 11.2 months, with the wetland being dry for only approximately 3 weeks in September. These changes in hydroperiod are likely due to variations in climate over the study period. 2016 and 2017 were both warmer than average years by 0.6°C and 0.8°C, respectively (National Oceanic and Atmospheric Administration, 2023). 2016 was also drier than average (4.8 cm less precipitation), while 2017 was wetter than average (13.3 cm greater precipitation) (National Oceanic and Atmospheric Administration, 2022). In contrast, 2018 and 2019 were both cooler than average, by 0.2°C and 0.9°C, respectively (National Oceanic and Atmospheric Administration, 2023) and had significantly greater than average precipitation (28.9 and 29.7 cm greater, respectively) (National Oceanic and Atmospheric Administration, 2023). These longer hydroperiods signal enhanced water retention during increasingly wet years and highlight the responsiveness of prairie wetlands to regional climate variability. These changes in hydroperiod also altered the internal hydrodynamics of the prairie by increasing surface connectivity (Figures 3 and 4 and associated discussion), and are expected to produce long-term changes in the prairie and wetland ecosystems, for example, facilitating displacement of native plant species but potentially providing greater habitat for nesting waterbirds (Donnelly et al., 2022; Foti et al., 2012; Miller & Zedler, 2003; Zedler & Kercher, 2004).
Hydrologic power spectra (Figure 7) exhibit a strong annual signal in precipitation, water level, and soil moisture throughout the prairie, as indicated by the peak at 365 days in each plot. In the precipitation signal, events are random aside from an annual cycle (i.e., wet and dry seasons), with a spectral signature consistent with white noise. Water level spectra show strong annual and daily signals. The daily signal can be attributed to evapotranspiration, which is present across the prairie from extensive vegetation cover, particularly in the summer months. The water level spectra yield 1/f pink noise with a log-log slope of approximately 2 in the high-intermediate frequency ranges (daily-monthly), demonstrating that surface-groundwater interactions cause the prairie to act as a fractal filter of the precipitation signal, as has been observed in stream catchments and soil moisture dynamics (Dwivedi et al., 2020; Kirchner et al., 2000; Kirchner & Neal, 2013). Most monitoring locations exhibit this behavior (Figures S3 and S4 in Supporting Information S1), including both groundwater and surface water, because all parts of the prairie are influenced by surface-groundwater interactions over a wide range of time scales. Notably, hydrologic fractal filtering was observed for locations with substantial surface-groundwater interactions (wetland and swale) and locations that are purely groundwater (sand ridge). The only significant deviations in the spectral scaling occurred in surface water in ditches along the northern edge of the prairie (WLS1, WLS3, and WLS8), which did not display a consistent power-law slope in the daily-to-monthly frequency range. These locations receive drainage from neighborhoods surrounding the prairie and are therefore influenced by hydrologic processes beyond the internal dynamics of the prairie. Near-surface soil moisture spectra are similar to those obtained from water levels, as expected. However, soil moisture spectra do not show as distinct an annual signal at depths of ∼1 m (Figure S3 in Supporting Information S1). At these deeper levels, soil is more consistently saturated and is only impacted by prolonged dry weather, notably in late summer 2017 (Figure 6). Together, these spectral signatures reflect how the prairie smooths and redistributes hydrologic signals through surface-subsurface interactions, acting as a dynamic buffer within the urban watershed.
[IMAGE OMITTED. SEE PDF]
Storm Event Dynamics
Hydrograph response is essential to understanding stormwater dynamics of the prairie-wetland system; infiltration and baseflow recession are also key components of the long-term site water balance and therefore the ecosystem response to changing seasonal climate. Hydrologic response parameters in the wetland (WLW1/WLW3/WLW6), sand ridge (WLW12) and swale (WLW14) are shown in Table 1 for six rainfall events. These storms were chosen because they induced a measurable response in water level across most regions of the prairie (whereas some storms only yielded a response in the wetland or swale but not the sand ridge) and they cover a wide range of antecedent soil conditions (Table S1 in Supporting Information S1). Across most storms, we observed shorter response and relaxation times in the sand ridge relative to the swale and wetland (Figure S6 in Supporting Information S1). This indicates more rapid infiltration into the high-permeability sand ridge than the clay underlying the wetland and swale. We observed essentially no correlation between the response and relaxation times and antecedent conditions, including soil moisture and recent (24 hr) precipitation (Table S2 in Supporting Information S1) across all of the well locations. This could be due to a limited range of antecedent conditions for which there were measurable responses across all of the regions of the prairie, as well as data gaps in the soil moisture measurements which further reduced available data.
Table 1 The Time to Peak, Relaxation Time, and Change in Water Level From Antecedent Level to Peak for Six Rain Events in the Wetland (WLW1/WLW3/WLW6), Sand Ridge (WLW12) and Swale (WLW14) Are Shown Below
| Rainfall event | Antecedent soil moisture (sand ridge; wetland) | Sensor name | Time to peak (hours) | Relaxation time (hours) | Initial water level (m) | Change in water level (m) |
| 8/18/19 (4.42 cm) | 13.5% ± 2.8; N/A | Sand Ridge | 7 | 41.5 | 0.05 | 0.13 |
| Swale | 4 | 38 | 0.57 | 0.04 | ||
| Wetland (WLW3) | 15 | 39.5 | 0.57 | 0.31 | ||
| 8/20/18 (3.73 cm) | 15.7% ± 2.9; 20.8% ± 7.8 | Sand Ridge | 13 | 177.5 | 0.06 | 0.23 |
| Swale | 16 | 180.5 | 0.31 | 0.66 | ||
| Wetland (WLW3) | 19 | 179 | 0.23 | 0.58 | ||
| 8/20/19 (1.64 cm) | 17.4% ± 3.5; N/A | Sand Ridge | 5.5 | 122 | 0.17 | 0.03 |
| Swale | 11.5 | 604.5 | 0.60 | 0.10 | ||
| Wetland (WLW3) | 9.5 | 134.5 | 0.75 | 0.15 | ||
| 7/21-7/22/17 (8.60 cm) | N/A; 22.4% ± 5.4 | Sand Ridge | 11.5 | 108.5 | 0.05 | 0.12 |
| Swale | -- | -- | -- | -- | ||
| Wetland (WLW3) | 7 | 381 | 0.32 | 0.64 | ||
| 2/19-2/21/18 (9.42 cm) | 18.2% ± 3.6; 29.2% ± 3.9 | Sand Ridge | 27.5 | 203 | 0.53 | 0.44 |
| Swale | -- | -- | -- | -- | ||
| Wetland (WLW6) | 18.5 | 212 | 0.69 | 0.25 | ||
| 6/20-6/22/18 (9.68 cm) | 32.0% ± 0.8; 41.3% ± 2.8 | Sand Ridge | 9 | 121 | 0.61 | 0.37 |
| Swale | -- | -- | -- | -- | ||
| Wetland (WLW1) | 20.5 | 304.5 | 0.16 | 0.75 |
Event hydrologic responses were modulated by seasonal dynamics. During the spring months, diurnal fluctuations in water levels were superimposed on the falling limb due to snowmelt during warmer, sunnier daytimes, compared to cold nights when temperatures return to below freezing. Diurnal fluctuations were also observed along the falling limb in the late summer during the peak growing season due to drawdown from ET accelerating the recession (Figure S7 in Supporting Information S1).
Stormwater Storage
Estimating the stormwater storage in GMP informs our understanding of how urban greenspaces and green infrastructure contribute to flood attenuation and reduction. We illustrate this by determining the storage volume for each measurement location for a rainfall event of 10.9 cm on 27 September 2019, which is essentially the same as the 10.92 cm of precipitation for the 5-year return period storm 24-hr duration (Angel & Markus, 2019). The average soil moisture in the 24 hr preceding the rain event was 16.9% ± 3.6% VWC in the top 40 cm of the sand ridge and 43.2% ± 1.2% in the top 40 cm of the wetland. The rainfall volume was estimated by multiplying the total rainfall by the area of each measurement location's assigned Voroni polygon from the discretization of the prairie surface. The water stored in the surface and subsurface for the September 27 rain event was approximately 77,100 m3, which is 64% greater than the estimated rainfall volume of 46,900 m3 that fell on the monitored portion of GMP (Figure S8 in Supporting Information S1). These results indicate that the prairie received and stored over 30,000 m3 of off-site stormwater that would otherwise have contributed to flooding in surrounding neighborhoods.
We extend this analysis to determine stormwater storage across the prairie for a range of antecedent soil moisture conditions. Results are shown in Figure 8 for two storms that are similar in magnitude but vary in season and antecedent conditions. Stormwater storage is calculated based on the catchment areas for each of the wells, as shown in Figure 8. The rainfall event on 19–21 February 2018 had an antecedent soil moisture of 29.2% ± 3.9% in the wetland and 18.2% ± 3.6% in the sand ridge and received 9.42 cm of rainfall (Figure 8a). The rainfall event on 20–22 June 2018 had an antecedent soil moisture of 41.3% ± 2.8% in the wetland and 32.0% ± 0.8% in the sand ridge and received 9.68 cm of rainfall (Figure 8b). Under wet antecedent conditions (Figure 8b), we observed less subsurface stormwater storage in the wetland areas and greater storage in the ditch and swale. This behavior is expected because greater soil moisture in the organic-rich clay soils underlying the wetland delays groundwater response and storage. The total storage provided by the prairie for the February 2018 event was approximately 15,650 m3, with 93.1% stored in the subsurface and only 6.9% stored as surface water. The total storage provided by the prairie for the June 2018 event was approximately 13,400 m3, with 94.6% stored in the subsurface and only 5.4% stored as surface water. These volumetric estimates provide insight into the prairie's retention capacity. However, they are based on porosity values derived from literature sources and are subject to an estimated uncertainty of ±10%–20%, due to natural variability in soil properties and the absence of direct measurements.
[IMAGE OMITTED. SEE PDF]
Despite this uncertainty, groundwater storage consistently accounts for the vast majority of total stormwater storage provided by the prairie under all antecedent conditions. This finding indicates that surface water-groundwater interactions control both the total stormwater storage capacity and the spatiotemporal distribution of water storage within the prairie-wetland system.
Approximately 413,000 m3 of precipitation falls onto GMP each year, and most of this water is retained within the site. Some surface water directly enters Bel Aire Creek via overland flow, while most enters the swale, wetland, or disconnected smaller ditches. Surface runoff from the adjacent neighborhood is also directed into ditches on the periphery of the prairie by the engineered urban drainage system of the surrounding communities. Bel Aire Creek, the primary surface flow channel exiting the prairie, was not observed to reach bankfull capacity during the study period. The sensor network data show that water surface elevations in Bel Aire Creek are typically lower than those in interior ditches and the wetland, and the hydrographs are flashier at both upstream and downstream locations in Bel Aire Creek than in any part of the prairie-wetland complex. There is rarely surface connectivity between the Bel Aire Creek and the wetland or interior ditches, because of the efforts of NEIU and TNC in blocking ditch connections to prevent dewatering of the prairie. Consequently, most water flow between GMP and Bel Aire Creek is through the subsurface. Nonetheless, the prairie receives and stores approximately 500,000 m3 annually, 87,000 m3 of which is estimated to come from the surrounding neighborhoods to the west and south, which would otherwise propagate rapidly into the Little Calumet River network and exacerbate the flooding that frequently occurs in downstream communities (notably Harvey and Robbins, IL). In comparison to an adjacent urban stream in the same watershed (The Calumet Union Drainage Canal, 2 km south of GMP; USGS Gauge 05536310) (United States Geological Survey, 2020), the prairie responds more slowly to precipitation and has smaller hydrograph peaks (Figure S9 in Supporting Information S1).
Discussion
GMP's hydrological dynamics are controlled by surface-groundwater interactions that vary seasonally. Interactions between surface water and groundwater within the prairie control both stormwater storage and hydrologic connectivity. Stormwater runoff is stored in GMP as surface water, groundwater, and soil water. Groundwater accounts for more than 80% of total stormwater storage across the prairie (Figure 8). The stormwater storage capacity of the prairie depends on the ability of water to infiltrate into soils and on surface and subsurface connectivity of the site, which is needed for water to reach locations of infiltration within relevant timescales. Long surface flow paths, from the upland area of the sand ridge to the swale, flatten event hydrographs and allow time for significant infiltration before stormwater reaches the wetland (Figures 2 and 4). These surface-groundwater interactions favor infiltration and long-term subsurface storage over a large area of the prairie relative to short-term stormwater storage in surface water. These findings are consistent with the broader understanding of hydrologic systems as fractal filters, where long and variable residence times in the subsurface lead to temporally persistent responses (Dwivedi et al., 2020; Kirchner et al., 2000).
Over longer timescales, groundwater flow from the sand ridge to the wetland is a slower process than event responses and overland flow (Pavlin et al., 2021), and affects seasonal dynamics by recharging the wetland. The wetland and swale regions have long-term water storage in both surface water and groundwater. The shorter observed seasonal hydroperiods of ∼5 months are typical of ephemeral wetlands and vernal pools, which provide important habitat for invertebrates, amphibians, and nesting waterfowl, all of which have been observed in GMP (McGuire, 2010). Extending wetland hydroperiods could potentially provide greater habitat for secretive marsh birds if emergent vegetation is still present (Malone et al., 2023). However, changes in the hydroperiod will also impact vegetation and can increase opportunities for invasive species to dominate the environment (Zedler & Kercher, 2004). Such changes in plant species subsequently impact habitat availability, nutrient structure, and food webs within the prairie-wetland ecosystem (Foti et al., 2012; Zedler & Kercher, 2004). Fluctuations in water levels have also been shown to control the microbial community structure and diversity across the prairie-wetland complex (Griffin et al., 2020). Convergent surface flow paths from the surrounding urban environment and within the prairie also recharge GMP's wetland system, but urban stormwater flows carry road salts that are stored seasonally and metals that accumulate within the aquatic system (Gonzalez et al., 2023; Hernandez Gonzalez et al., 2019).
Infiltration in upland soils of the sand ridge and the area northeast of the swale provides short-term water storage via soil wetting accounting for approximately 34%–39% of the total groundwater storage provided by the prairie and reduces runoff from high-intensity storm events. In wet seasons, the prairie soils are saturated near the surface for 6–7 months of the year, preventing further infiltration and routing the bulk of precipitation to overland flow. Subsurface soil storage capacity is more limited in the wetter months, when there is also less evapotranspiration to remove near-surface water. This impacts both the available type and capacity of stormwater storage. Under dry antecedent conditions (e.g., February 2018 rainfall event), the prairie had approximately 17% greater storage than under wet antecedent conditions (June 2018 rainfall event). Conversely, when antecedent subsurface soil moisture is elevated, up to 30% more water may be stored as surface water in wetlands and ditches. However, under wet conditions, there is less storage overall, as much less groundwater storage is available, making surface water storage more important.
Surface-subsurface hydrological dynamics within GMP also play an important role in regulating downstream stormwater discharge. For a 5-year, 24-hr storm in this region (∼10.9 cm rainfall), the prairie stores ∼30,200 m3 of stormwater derived from the surrounding environment under dry antecedent conditions and a total storage density of 1,930 m3/ha. This storage per area estimate closely matches storage estimates by Lane & D’Amico, 2010, who calculated an average storage per area of 1,619 m3/ha for isolated wetlands in Florida (Lane & D’Amico, 2010). It is worth noting that these values are significantly lower than estimates of storage per area for prairie potholes in Minnesota of approximately 14,000 m3/ha (Gleason et al., 2007). However, prairie potholes often have greater depth relative to area, which may allow them to store more water per unit surface area than flatter prairie wetlands like those at GMP. This storage is modulated by seasonal variations and antecedent conditions, but nonetheless persistently reduces stormwater discharge that contributes to flooding. The slower response and smaller hydrograph peaks of the prairie in comparison to an adjacent urban stream (Figure S9 in Supporting Information S1) are important because GMP is part of the headwaters of the Little Calumet River watershed, which suffers considerable flooding damage downstream in the Calumet area of NE Illinois. Based on the results presented here and assuming similar hydrologic responses for the other prairies, we estimate that the entire Indian Boundary Prairies complex—which comprises five natural and restored prairies with a total area of 189 ha—provides regional storage of approximately 234,000 m3 per year of stormwater derived from the surrounding environment and protects an array of highly flood-impacted communities in Southern Cook County. However, this extrapolation is not based on direct measurements or modeling across all IBP sites so it should be interpreted as a preliminary, illustrative estimate.
Stormwater storage and retention at GMP are most effective during drier periods when the soil is not fully saturated, and does not provide as much storage under wet antecedent conditions when there are multiple rainfall events in a short period of time. This has important implications for the efficacy of highly vegetated nature-based solutions to reduce seasonal flooding in the region. The prairie provides less storage in the spring and early summer when the groundwater table and soil moisture are high. In the summer and fall, when the groundwater table is lower and soils are drier, the prairie provides greater storage and can more effectively reduce downstream stormwater discharge. Regional flooding occurs throughout these seasons owing to the prevalence of unstable atmospheric conditions from jet stream fluctuations in the spring and fall, and the propensity to intense convective thunderstorms in the summer (Changnon & Kunkel, 2006).
Regional climate change is expected to alter internal prairie ecohydrology and exacerbate flooding in the surrounding urban environment. Precipitation has increased by 15% in the past century, and winters have become wetter and warmer (Hayhoe et al., 2010; U.S. Global Change Research Program, 2018; Wuebbles et al., 2021; Jay et al., 2023). These trends are expected to continue across the Midwest and Great Lakes regions. In addition, precipitation events are expected to become less frequent and more intense in the future (Jay et al., 2023; Wuebbles et al., 2021). These conditions are expected to cause more rain-on-snow events, which yield much higher runoff compared to event precipitation and produce outsized flooding impacts (Li et al., 2019). Warmer winter conditions and increased total precipitation are extending the wetland hydroperiod. We observed that hydroperiods in GMP increased from 4.6 months in 2017 to 7.6 months in 2018 and 11.2 months in 2019. 2018 and 2019 experienced greater than average rainfall, with 2019 the fifth wettest year on record in Illinois (Wuebbles et al., 2021). Rain-on-snow events occurred frequently during this period: 21 between December 2017–March 2018 and 26 between November 2018-February 2019. These wetter winter conditions are expected to occur more frequently, and the associated hydrologic changes threaten both the exceptional prairie-wetland ecosystem in GMP and broader regional ecosystems that depend on prairie and wetland habitat (Miller & Zedler, 2003; Zedler & Kercher, 2004). In addition, expected changes in summer conditions, with less frequent and more intense precipitation, can impact prairie ecosystems by reducing vegetation production (Yongguang et al., 2013). These ecosystems are already fragile, hosting multiple endangered and threatened species that currently have very limited habitat because of extensive land conversion across the central U.S. The combination of decreased rainfall frequency and increased intensity is expected to make the prairie swale/wetland system more ephemeral in the summer, while warmer and wetter winters will shift the hydroperiod earlier in the year. More intense storms will result in rapid filling of swale and wetland, leading to greater surface connectivity across the site, even if the water is not present for long periods of time.
Building on our site-scale observations of stormwater storage and hydrologic response, broader insights from socio-hydrology emphasize that the effectiveness of natural systems for flood mitigation is closely linked to the dynamic interactions between land use, hydrologic processes, and human decision-making. Attaran et al. (2024) and Pan et al. (2018) demonstrate that urbanization and drainage modifications can strongly influence flood risk, often more than changes in climate or precipitation alone. This highlights the need to preserve existing ecologically functional landscapes within urban regions. Gohari et al. (2022) further emphasize that integrated modeling of water-land-society systems can help anticipate the cumulative impacts of these interactions and support adaptive management. Applying these perspectives to urban prairies and wetlands underscores their role not only as ecological resources, but also as critical infrastructure for long-term resilience in the face of changing climate and land use conditions.
While this study is geographically constrained, it provides a detailed, process-based understanding of ecohydrological dynamics in embedded natural systems that can inform conceptual models of stormwater behavior in the context of other urban prairies or green infrastructure. Many of the processes observed here, such as the dominance of subsurface storage, the importance of surface-subsurface connectivity, and the role of antecedent conditions, are fundamental hydrological properties that are likely to apply across a range of urban green infrastructure and remnant natural systems. These findings highlight the potential for native prairie and wetland ecosystems to serve as multifunctional green infrastructure, offering both stormwater storage and biodiversity benefits. Incorporating such natural systems into regional flood resilience planning can enhance climate adaptation while preserving ecological value. Restoration and conservation efforts that prioritize hydrologic connectivity and native vegetation can improve both water management and habitat outcomes across urbanizing landscapes.
Conclusions
This study demonstrates that natural areas embedded in the urban environment, like prairies and wetlands, provide substantial storage of stormwater originating off-site. Based on long-term, high-frequency data from a network of surface and groundwater sensors, we determined that Gensburg-Markham Prairie stores approximately 87,000 m3 of off-site stormwater annually, acknowledging uncertainty in porosity estimates that impact volumetric calculations. Notably, we found that stormwater storage within the prairie-wetland complex is driven primarily by surface-groundwater interactions, and both short- and long-term stormwater storage occurs predominantly via fluctuations in the groundwater table. Spectral scaling analysis revealed that these site-scale surface-groundwater interactions produce fractal filtering of the hydrologic time-series over daily-to-monthly timescales, similar to behavior observed previously for river-groundwater interactions and at reach-to-watershed scales.
Stormwater storage is greatest when the prairie soil and wetland are dry prior to a precipitation event. Stormwater storage increases by approximately 17% under dry antecedent conditions relative to wet antecedent conditions. Wetter conditions suppress surface-groundwater interactions, which reduces infiltration and storage of stormwater in the subsurface. Consequently, stormwater storage is more limited under wet antecedent conditions, with more storage in wetland surface water relative to groundwater, greater connectivity of the internal surface water flow system, and more rapid export to the downstream urban surface water network. These results show that surface-subsurface connectivity is critical to achieving high ecosystem value for flood protection, while high surface flow connectivity favors runoff to downstream urban areas.
This understanding of the ecohydrological dynamics of embedded natural areas informs the role urban green spaces play in flood reduction. However, extrapolation to other areas will require consideration of local characteristics, including vegetation, soil properties, and hydrologic connectivity. In particular, our analysis of the relationship between ecohydrological characteristics and stormwater storage illustrates the total stormwater storage provided by urban prairies and wetlands, the critical importance of surface-groundwater interactions to flood protection, and how climate and other seasonal conditions influence these outcomes. The methods described in this paper can readily be extended to other types of urban greenspaces and used to estimate stormwater storage provided by embedded ecosystems at the regional scale and improve the design of nature-based solutions for regional stormwater management.
Acknowledgments
The authors thank John Legge, Matt Mulligan, and Shane Tripp at The Nature Conservancy and Jennifer Slate and Anthony Merisko at Northeastern Illinois University for facilitating this research at GMP and sharing their expertise. This material is based upon work supported by the U.S. Department of Energy, Office of Science, Office of Biological and Environmental Research’s Urban Integrated Field Laboratories CROCUS project under Contract Number DE-AC02-06CH11357 and Award Number DE-SC0023364, and the U.S. National Science Foundation under awards 1848683 and 1324585.
Data Availability Statement
Water level data from Gensburg-Markham Prairie have been made available using Hydroshare (O’Brien et al., 2024).
American Society for Testing and Materials. (2020). D2487 standard practice for classification of soils for engineering purposes. Unified Soil Classification System). Retrieved from https://www.astm.org/d2487‐17.html
Amoozegar, A., Heitman, J. L., & Kranz, C. N. (2023). Comparison of soil particle density determined by a gas pycnometer using helium, nitrogen, and air. Soil Science Society of America Journal, 87(1), 1–12. https://doi.org/10.1002/saj2.20476
Ando, A., Hanson, A.‐M., Kocs, E., Klein‐Banai, C., Massey, D., Rigsbee, C., et al. (2020). 2019 CURES connections workshop: New voices and paths to urban sustainability. Retrieved from http://hdl.handle.net/2142/109993
Angel, J., & Markus, M. (2019). Frequency distributions of heavy precipitation in Illinois: Updated bulletin 70. ISWS contract report CR‐2019‐05. Retrieved from https://www.ideals.illinois.edu/items/110212
Attaran, S., Mosaedi, A., Sojasi Qeidari, H., & Farrokhi Derakhshandeh, J. (2024). Co‐evolution of human and hydrological system: Presenting a socio‐hydrological approach to flood adaptation in Kalat city, Iran. International Journal of Disaster Risk Reduction, 102, 104292. https://doi.org/10.1016/j.ijdrr.2024.104292
Beaubien, G. (1994). “Living history.” the Chicago Tribune. Oct 13, 1994. Retrieved from http://www.chicagotribune.com/news/ct‐xpm‐1994‐10‐13‐9410130296‐story.html
Beckman, P., Sankaran, R., Catlett, C., Ferrier, N., Jacob, R., & Papka, M. (2016). Waggle: An open sensor platform for edge computing. In 2016 IEEE SENSORS (pp. 1–3). https://doi.org/10.1109/ICSENS.2016.7808975
Beniston, M., & Stoffel, M. (2016). Rain‐on‐snow events, floods and climate change in the Alps: Events may increase with warming up to 4°C and decrease thereafter. Science of the Total Environment, 571, 228–236. https://doi.org/10.1016/j.scitotenv.2016.07.146
Bhaskar, A. S., Hogan, D. M., Nimmo, J. R., & Perkins, K. S. (2018). Groundwater recharge amidst focused stormwater infiltration. Hydrological Processes, 32(13), 2058–2068. https://doi.org/10.1002/hyp.13137
Borden, K., Schmidtlein, M., Emrich, C., Piegorsch, W., & Cutter, S. (2007). Vulnerability of U.S. cities to environmental hazards. Journal of Homeland Security and Emergency Management, 4(2). https://doi.org/10.2202/1547‐7355.1279
Carsel, R. F., & Parrish, R. S. (1988). Developing joint probability distributions of soil water retention characteristics. Water Resources Research, 24, 5–769. https://doi.org/10.1029/WR024i005p00755
Center for Neighborhood Technology. (2014). The prevalence and cost of urban flooding. Center for Neighborhood Technology. Retrieved from https://cnt.org/publications/the‐prevalence‐and‐cost‐of‐urban‐flooding
Changnon, S., & Kunkel, K. (2006). Severe storms in the midwest. Illinois State Water Survey. Retrieved from https://www.isws.illinois.edu/pubdoc/iem/iswsiem2006‐06.pdf
Chicago Metropolitan Agency for Planning. (2017). Strategies to reduce the effects of flooding in metropolitan Chicago. Retrieved from https://www.cmap.illinois.gov/updates/all/‐/asset_publisher/UIMfSLnFfMB6/content/strategies‐to‐reduce‐the‐effects‐of‐flooding‐in‐metropolitan‐chicago
Chicago Metropolitan Agency for Planning. (2018). Guide to flood susceptibility and stormwater planning. Retrieved from https://www.cmap.illinois.gov/web/guest/programs/water/stormwater/stormwater‐and‐flooding
Climate Engine. (2021). Desert Research Institute and University of Idaho, version 2.1. Retrieved from http://climateengine.org
Cook County Bureau of Technology, GIS Division. (2023). Cook county 2022 LiDAR and topographic data services, 6‐inch contours for cook county, Illinois. Geodatabase Vector Digital Data. Published: June 2023.
Dahl, T. E. (1990). Wetlands losses in the United States, 1780's to 1980's. Report to the Congress (PB‐91‐169284/XAB). National Wetlands Inventory. Retrieved from https://www.osti.gov/biblio/5527872
Dahl, T. E. (2011). Status and trends of wetlands in the conterminous United States 2004 to 2009.
Davenport, F. V., Burke, M., & Diffenbaugh, N. S. (2021). Contribution of historical precipitation change to US flood damages. Proceedings of the National Academy of Sciences of the United States of America, 118(4), e2017524118. https://doi.org/10.1073/pnas.2017524118
Davidson, N. C. (2014). How much wetland has the world lost? Long‐term and recent trends in global wetland area. Marine and Freshwater Research, 65(10), 934–941. https://doi.org/10.1071/mf14173
Donnelly, J. P., Moore, J. N., Casazza, M. L., & Coons, S. P. (2022). Functional wetland loss drives emerging risks to waterbird migration networks. Frontiers in Ecology and Evolution, 10, 844278. https://doi.org/10.3389/fevo.2022.844278
Dunn, J. B., & Miller, W. M. (2020). Sustainable urban systems: Predictive, interconnected, resilient, and evolving. https://doi.org/10.21985/n2‐3mvx‐m804
Dwivedi, R., Knowles, J. F., Eastoe, C., Minor, R., Abramson, N., Mitra, B., et al. (2020). Ubiquitous fractal scaling and filtering behavior of hydrologic fluxes and storages from a mountain headwater catchment. Water, 12(2), 613. Article 2. https://doi.org/10.3390/w12020613
Eaton, T. T. (2018). Approach and case‐study of green infrastructure screening analysis for urban stormwater control. Journal of Environmental Management, 209, 495–504. https://doi.org/10.1016/j.jenvman.2017.12.068
Ford, T. W., Chen, L., & Schoof, J. T. (2021). Variability and transitions in precipitation extremes in the midwest United States. Journal of Hydrometeorology, 22(3), 533–545. https://doi.org/10.1175/JHM‐D‐20‐0216.1
Foti, R., del Jesus, M., Rinaldo, A., & Rodriguez‐Iturbe, I. (2012). Hydroperiod regime controls the organization of plant species in wetlands. Proceedings of the National Academy of Sciences of the United States of America, 109(48), 19596–19600. https://doi.org/10.1073/pnas.1218056109
Gleason, R. A., Tangen, B. A., Laubhan, M. K., Kermes, K. E., & Euliss, N. H. (2007). Estimating water storage capacity of existing and potentially restorable wetland depressions in a subbasin of the Red River of the North. ResearchGate. Retrieved from https://www.researchgate.net/publication/301312767_Estimating_Water_Storage_Capacity_of_Existing_and_Potentially_Restorable_Wetland_Depressions_in_a_Subbasin_of_the_Red_River_of_the_North
Gohari, A., Savari, P., Eslamian, S., Etemadi, N., & Keilmann‐Gondhalekar, D. (2022). Developing a system dynamic plus framework for water‐land‐society nexus modeling within urban socio‐hydrologic systems. Technological Forecasting and Social Change, 185, 122092. https://doi.org/10.1016/j.techfore.2022.122092
Golden, H. E., & Hoghooghi, N. (2018). Green infrastructure and its catchment‐scale effects: An emerging science. WIREs. Water, 5(1), 1254. https://doi.org/10.1002/wat2.1254
Gonzalez, L. M. H., Rivera, V. A., Akosa, D., Phillips, C. B., Hatch, S. L., Miller, W. M., & Packman, A. I. (2023). Road salt intrusion dynamics in an ex‐urban native wetland complex. PLOS Water, 2(7), e0000148. https://doi.org/10.1371/journal.pwat.0000148
Green, D., O'Donnell, E., Johnson, M., Slater, L., Thorne, C., Zheng, S., et al. (2021). Green infrastructure: The future of urban flood risk management? WIREs Water, 8(6), e1560. https://doi.org/10.1002/wat2.1560
Griffin, J. S., Haug, L. A., Rivera, V. A., Gonzalez, L. M. H., Kelly, J. J., Miller, W. M., et al. (2020). Soil hydrology drives ecological niche differentiation in a native prairie microbiome. FEMS Microbiology Ecology, 96(1), fiz163. https://doi.org/10.1093/femsec/fiz163
Hanson, P. (1975). A vegetational study of the Gensburg Markham prairie. Northeastern Illinois University.
Hayhoe, K., VanDorn, J., Croley, T., Schlegal, N., & Wuebbles, D. (2010). Regional climate change projections for Chicago and the US Great Lakes. Journal of Great Lakes Research, 36, 7–21. https://doi.org/10.1016/j.jglr.2010.03.012
Hernandez Gonzalez, L. M., Rivera, V. A., Phillips, C. B., Haug, L. A., Hatch, S. L., Yeager, L. E., et al. (2019). Characterization of soil profiles and elemental concentrations reveals deposition of heavy metals and phosphorus in a Chicago‐area nature preserve, Gensburg Markham Prairie. Journal of Soils and Sediments, 19(11), 3817–3831. https://doi.org/10.1007/s11368‐019‐02315‐5
Hole, F. D., & Nielsen, G. A. (1968). Soil genesis under prairie. In Proceedings of a Symposium on Prairie and Prairie Restoration (pp. 28–35). Knox College, Galesburg.
Hu, S., Niu, Z., Chen, Y., Li, L., & Zhang, H. (2017). Global wetlands: Potential distribution, wetland loss, and status. The Science of the Total Environment, 586, 319–327. https://doi.org/10.1016/j.scitotenv.2017.02.001
Hunt, A., & Watkiss, P. (2011). Climate change impacts and adaptation in cities: A review of the literature. Climatic Change, 104(1), 13–49. https://doi.org/10.1007/s10584‐010‐9975‐6
Huntington, J., Hegewisch, K., Daudert, B., Morton, C., Abatzoglou, J., McEvoy, D., & Erickson, T. (2017). Climate engine: Cloud computing of climate and remote sensing data for advanced natural resource monitoring and process understanding. Bulletin of the American Meteorological Society, 98(11), 2397–2410. https://doi.org/10.1175/BAMS‐D‐15‐00324.1
Illinois Department of Natural Resources. (n.d.). Schwegman's natural divisions of Illinois. Illinois Department of Natural Resources. Retrieved from https://dnr.illinois.gov/education/atoz/schwegman.html
Illinois State Water Survey. (n.d.). Climate of Chicago. Retrieved from https://www.isws.illinois.edu/statecli/general/chicago‐climate‐narrative.htm
Jay, A. K., Crimmins, A. R., Avery, C. W., Dahl, T. A., Dodder, R. S., Hamlington, B. D., et al. (2023). Fifth national climate assessment. U.S. Global Change Research Program. https://doi.org/10.7930/NCA5.2023.CH1
Keenan, M. B., Shankar, P., & Haas, P. (2020). Assessing disparities of urban flood risk for households of color in Chicago. Center for Neighborhood Technology. Retrieved from https://cnt.org/publications/assessing‐disparities‐of‐urban‐flood‐risk‐for‐households‐of‐color‐in‐chicago
Kirchner, J. W., Feng, X., & Neal, C. (2000). Fractal stream chemistry and its implications for contaminant transport in catchments. Nature, 403(6769), 524–527. https://doi.org/10.1038/35000537
Kirchner, J. W., & Neal, C. (2013). Universal fractal scaling in stream chemistry and its implications for solute transport and water quality trend detection. Proceedings of the National Academy of Sciences of the United States of America, 110(30), 12213–12218. https://doi.org/10.1073/pnas.1304328110
Kucharik, C. (2007). Impact of prairie age and soil order on carbon and nitrogen sequestration. Soil Science Society of America Journal, 71(2), 430–441. Retrieved from https://acsess‐onlinelibrary‐wiley‐com.turing.library.northwestern.edu/doi/full/10.2136/sssaj2006.0074
Lane, C. R., & D'Amico, E. (2010). Calculating the ecosystem service of water storage in isolated wetlands using LiDAR in North Central Florida, USA. Wetlands, 30(5), 967–977. https://doi.org/10.1007/s13157‐010‐0085‐z
Larsen, T. A., Hoffmann, S., Lüthi, C., Truffer, B., & Maurer, M. (2016). Emerging solutions to the water challenges of an urbanizing world. Science, 352(6288), 928–933. https://doi.org/10.1126/science.aad8641
Li, D., Lettenmaier, D. P., Margulis, S. A., & Andreadis, K. (2019). The role of rain‐on‐snow in flooding over the conterminous United States. Water Resources Research, 55(11), 8492–8513. https://doi.org/10.1029/2019WR024950
Malone, K. M., Webb, E. B., Mengel, D. C., Kearns, L. J., McKellar, A. E., Matteson, S. W., & Williams, B. R. (2023). Wetland management practices and secretive marsh bird habitat in the Mississippi Flyway: A review. Journal of Wildlife Management, 87(7), e22451. https://doi.org/10.1002/jwmg.22451
Maragno, D., Gaglio, M., Robbi, M., Appiotti, F., Fano, E. A., & Gissi, E. (2018). Fine‐scale analysis of urban flooding reduction from green infrastructure: An ecosystem services approach for the management of water flows. Ecological Modelling, 386, 1–10. https://doi.org/10.1016/j.ecolmodel.2018.08.002
McGuire, C. (2010). Sustainable approaches to managing small‐scale ecosystems: A case study of vernal pool protection in the commonwealth of Massachusetts, United States of America (SSRN Scholarly Paper 2958227). Retrieved from https://papers.ssrn.com/abstract=2958227
Metropolitan Planning Council. (2020). StormStoreTM. Metropolitan Planning Council. Retrieved from https://www.metroplanning.org/work/project/48
Metropolitan Water Reclamation District of Greater Chicago. (2017). Stormwater management program 2017 annual report. Retrieved from https://mwrd.org/sites/default/files/documents/Stormwater%20Management_2017_Annual_Report.pdf
Miller, R. C., & Zedler, J. B. (2003). Responses of native and invasive wetland plants to hydroperiod and water depth. Plant Ecology, 167(1), 57–69. https://doi.org/10.1023/A:1023918619073
National Oceanic and Atmospheric Administration. (2022). Annual precipitation rankings for Chicago, Illinois. NOAA's National Weather Service. Retrieved from https://www.weather.gov/lot/Annual_Precip_Rankings_Chicago
National Oceanic and Atmospheric Administration. (2023). Annual temperature rankings for Chicago. NOAA's National Weather Service. Retrieved from https://www.weather.gov/lot/Annual_Temperature_Rankings_Chicago
National Operational Hydrologic Remote Sensing Center. (2004). Snow Data Assimilation System (SNODAS) data products at NSIDC, version 1 [Dataset]. National Snow and Ice Data Center. https://doi.org/10.7265/N5TB14TC
O'Brien, C., Rivera, V. A., Hernandez, L. M., Phillips, C. B., Gnaedinger, K., Pfeiffer, E., et al. (2024). Water level measurements; Gensburg Markham Prairie; 2016–2020 [Dataset]. Hydro. https://doi.org/10.4211/hs.6bebe63a3d564ecbaa78b8919260b3c0
Pan, H., Deal, B., Destouni, G., Zhang, Y., & Kalantari, Z. (2018). Sociohydrology modeling for complex urban environments in support of integrated land and water resource management practices. Land Degradation & Development, 29(10), 3639–3652. https://doi.org/10.1002/ldr.3106
Pavlin, L., Széles, B., Strauss, P., Paul Blaschke, A., & Blöschl, G. (2021). Event and seasonal hydrologic connectivity patterns in an agricultural headwater catchment. Hydrology and Earth System Sciences, 25(4), 2327–2352. https://doi.org/10.5194/hess‐25‐2327‐2021
Perktold, J., Skipper, S., Sheppard, K., ChadFulton, K. S., Jbrockmendel, J.‐G. 6, Quackenbush, P., et al. (2023). statsmodels/statsmodels: Release 0.14.0 (v0.14.0). Zenodo. https://doi.org/10.5281/ZENODO.593847
Rawls, W. J., Brakensiek, D. L., & Saxtonn, K. E. (1982). Estimation of soil water properties. Transactions of the ASAE, 25(5), 1316–1320. https://doi.org/10.13031/2013.33720
Rizzo, A., Bresciani, R., Masi, F., Boano, F., Revelli, R., & Ridolfi, L. (2018). Flood reduction as an ecosystem service of constructed wetlands for combined sewer overflow. Journal of Hydrology, 560, 150–159. https://doi.org/10.1016/j.jhydrol.2018.03.020
Shjeflo, J. B. (1968). Evapotranspiration and the water budget of prairie potholes in North Dakota. In Professional Paper (585‐B). U.S. Government Printing Office. https://doi.org/10.3133/pp585B
Soil Survey Staff, Natural Resources Conservation Service, United States Department of Agriculture. (2015). Soil Survey Geographic (SSURGO) database. Available online at https://sdmdataaccess.sc.egov.usda.gov
Surfleet, C. G., & Tullos, D. (2013). Variability in effect of climate change on rain‐on‐snow peak flow events in a temperate climate. Journal of Hydrology, 479, 24–34. https://doi.org/10.1016/j.jhydrol.2012.11.021
Tafasca, S., Ducharne, A., & Valentin, C. (2020). Weak sensitivity of the terrestrial water budget to global soil texture maps in the ORCHIDEE land surface model. Hydrology and Earth System Sciences, 24(7), 3753–3774. https://doi.org/10.5194/hess‐24‐3753‐2020
Thorp, J. (1948). In A. Stefferud (Ed.), How soils develop under grass. Grass: The yearbook of agriculture (pp. 55–66).
United States Department of the Army. (2012). Military Soils Engineering. Retrieved from https://armypubs.army.mil/epubs/DR_pubs/DR_a/pdf/web/tm3_34x64.pdf
United States Geological Survey. (2020). USGS current conditions for USGS 05536310 CALUMET UNION DRAINAGE CANAL NEAR MARKHAM, IL. Retrieved from https://nwis.waterdata.usgs.gov/nwis/dv?cb_00065=on&format=gif_default&site_no=05536310&legacy=&referred_module=sw&period=&begin_date=2020‐01‐09&end_date=2020‐01‐13
US Army Corps of Engineers, Rock Island District. (n.d.). Prairies of the Midwest. US Army Corps of Engineers. Retrieved from https://www.mvr.usace.army.mil/Portals/48/docs/Recreation/ODM/pdf/Prairies%20of%20the%20Midwest.pdf
U.S. Census Bureau. (2020). QuickFacts—Markham, Illinois. Retrieved from https://www.census.gov/quickfacts/markhamcityillinois
U.S. Global Change Research Program. (2018). Fourth national climate assessment (pp. 1–470). U.S. Global Change Research Program. Retrieved from https://nca2018.globalchange.gov/chapter/21
VanderPlas, J. T. (2018). Understanding the lomb‐scargle periodogram. The Astrophysical Journal ‐ Supplement Series, 236(1), 16. https://doi.org/10.3847/1538‐4365/aab766
Voter, C. B., & Loheide, S. P., II. (2018). Urban residential surface and subsurface hydrology: Synergistic effects of low‐impact features at the parcel scale. Water Resources Research, 54(10), 8216–8233. https://doi.org/10.1029/2018WR022534
Wagle, P., Xiao, X., Gowda, P., Basara, J., Brunsell, N., Steiner, J., & Anup, K. C. (2017). Analysis and estimation of tallgrass prairie evapotranspiration in the central United States. Agricultural and Forest Meteorology, 232, 35–47. https://doi.org/10.1016/j.agrformet.2016.08.005
Wu, Q., & Lane, C. R. (2017). Delineating wetland catchments and modeling hydrologic connectivity using lidar data and aerial imagery. Hydrology and Earth System Sciences, 21(7), 3579–3595. https://doi.org/10.5194/hess‐21‐3579‐2017
Wuebbles, D., Angel, J., Petersen, K., & Lemke, A. M. (Eds.) (2021). An assessment of the impacts of climate change in Illinois. In The Nature Conservancy. https://doi.org/10.13012/B2IDB‐1260194_V1
Wuebbles, D., Meehl, G., Hayhoe, K., Karl, T. R., Kunkel, K., Santer, B., et al. (2014). CMIP5 climate model analyses: Climate extremes in the United States. Bulletin of the American Meteorological Society, 95(4), 571–583. https://doi.org/10.1175/BAMS‐D‐12‐00172.1
Yongguang, Z., Moran, S., Nearing, M., Ponce Campos, G. E., Huete, A., Buda, A., et al. (2013). Extreme precipitation patterns and reductions of terrestrial ecosystem production across biomes. Journal of Geophysical Research: Biogeosciences, 118(1), 148–157. https://doi.org/10.1029/2012JG002136
Zedler, J. B., & Kercher, S. (2004). Causes and consequences of invasive plants in wetlands: Opportunities, opportunists, and outcomes. Critical Reviews in Plant Sciences, 23(5), 431–452. https://doi.org/10.1080/07352680490514673
Zedler, J. B., & Kercher, S. (2005). Wetland resources: Status, trends, ecosystem services, and restorability. Annual Review of Environment and Resources, 30(1), 39–74. https://doi.org/10.1146/annurev.energy.30.050504.144248
© 2025. This work is published under http://creativecommons.org/licenses/by/4.0/ (the "License"). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.