Journal of Environmental Management 362 (2024) 121284

Contents lists available at ScienceDirect

Journal of Environmental Management

journal homepage: www.elsevier.com/locate/jenvman

Research article

Impacts of future climate and land use/land cover change on urban runoff
using fine-scale hydrologic modeling
Lauren Ashley Mayou a, b, Nasrin Alamdari a, b, Ebrahim Ahmadisharaf a, b, *, Meysam Kamali c
a
Department of Civil and Environmental Engineering, FAMU-FSU College of Engineering, Tallahassee, FL, 32310, USA
b
Resilient Infrastructure and Disaster Response Center, FAMU-FSU College of Engineering, Tallahassee, FL, 32310, USA
c
Environment and Water Research Center, Sharif University of Technology, Tehran, Iran

A R T I C L E I N F O A B S T R A C T

Handling Editor: Jason Michael Evans Future changes in land use/land cover (LULC) and climate (CC) affect watershed hydrology. Despite past
research on estimating such changes, studies on the impacts of both these nonstationary stressors on urban
Keywords: watersheds have been limited. Urban watersheds have several important details such as hydraulic infrastructure
Climate change (CC) that call for fine-scale models to predict the impacts of LULC and CC on watershed hydrology. In this paper, a
Land use/land cover (LULC) change
fine-scale hydrologic model—Personal Computer Storm Water Management Model (PCSWMM)—was applied to
Urban watersheds
predict the individual and joint impacts of LULC changes and CC on surface runoff attributes (peak and volume)
Surface runoff
Hydrologic modeling in 3800 urban subwatersheds in Midwest Florida. The subwatersheds a range of characteristics in terms of
drainage area, surface imperviousness, ground slope and LULC distribution. The PCSWMM also represented
several hydraulic structures (e.g., ponds and pipes) across the subwatersheds. We analyzed changes in the runoff
attributes to determine which stressor is most responsible for the changes and what subwatersheds are mostly
sensitive to such changes. Six 24-h design rainfall events (5- to 200-year recurrence intervals) were studied under
historical (2010) and future (year 2070) climate and LULC. We evaluated the response of the subwatersheds in
terms of runoff peak and volume to the design rainfall events using the PCSWMM. The results indicated that,
overall, CC has a greater impact on the runoff attributes than LULC change. We also found that LULC and climate
induced changes in runoff are generally more pronounced in greater recurrence intervals and subwatersheds with
smaller drainage areas and milder slopes. However, no relationship was found between the changes in runoff and
original subwatershed imperviousness; this can be due to the small increase in urban land cover projected for the
study area. This research helps urban planners and floodplain managers identify the required strategies to protect
urban watersheds against future LULC change and CC.

1. Introduction The complexity of urban watersheds, characterized by their intricate

hydraulic infrastructure and diverse land management practices, de­
Land use/land cover (LULC) change due to urbanization results in mands advanced hydrologic modeling approaches (Tague and
more impervious areas and subsequently leads to smaller infiltration Pohl-Costello, 2008). Traditional coarse-scale models often fall short in
rates and greater runoff peaks and volumes (Alamdari et al., 2022; Huq capturing the detailed nuances of urban landscapes, particularly when
and Abdul-Aziz, 2021; Jarnagin, 2007; Moniruzzaman et al., 2020). assessing the combined impacts of LULC change and CC. Urban envi­
Climate change (CC) is projected to increase the frequency and intensity ronments, with their high degree of impervious surfaces, altered
of rainfall events, thereby affecting watershed hydrology (Fowler et al., drainage networks and unique hydrologic processes, require a more
2021; Willems et al., 2012). Frequent high-intensity rainfall events nuanced modeling approach (Ndulue et al., 2015). This is especially
result in increased runoff volume, flooding and damages to local com­ relevant when considering future scenarios of urban expansion and
munities and civil infrastructure (Arnbjerg-Nielsen et al., 2013). Com­ intensified weather events.
bined together, these nonstationary stressors—LULC change and Several studies have evaluated the impact of these nonstationary
CC—substantially alter watershed hydrology. changes on watershed hydrology (Nasta et al., 2017; (Q. Chen et al.,

* Corresponding author. 1753 W. Paul Dirac Dr, Tallahassee, FL, 32310, USA.
E-mail addresses: [email protected], [email protected] (E. Ahmadisharaf).

https://doi.org/10.1016/j.jenvman.2024.121284
Received 16 January 2024; Received in revised form 27 May 2024; Accepted 27 May 2024
Available online 4 June 2024
0301-4797/© 2024 Elsevier Ltd. All rights are reserved, including those for text and data mining, AI training, and similar technologies.
L.A. Mayou et al. Journal of Environmental Management 362 (2024) 121284

2019; Hansen et al., 2021; Hovenga et al., 2016; Marhaento et al., 2018; in this study sets a new standard for urban watershed analyses, offering
Ni et al., 2021; Rajaei et al., 2021; Sadeghi et al., 2020; Suttles et al., insights that are more relevant for urban planners. Our paper provides
2018; Villamizar et al., 2019; Wang et al., 2014; Wilson, 2015). These insights about plausible changes in urban runoff under nonstationary
studies have utilized hydrologic models such as Soil & Water Assessment LULC and climate in the future and supports planners and decision
Tool (SWAT (Neitsch et al., 2011);), Hydrologic Simulation Pro­ makers in proposing mitigation and adaptation strategies.
gram—FORTRAN (HSPF; Bicknell et al., 2005) and SPAtially Referenced
Regression on Watershed attributes (SPARROW; Schwarz et al., 2006) 2. Methodology
and Storm Water Management Model (SWMM; (Rossman, 2015). There
has been also studies that evaluated the impact of CC on watershed A hydrologic model—PCSWMM—was used to simulate hydrologic
hydrology (Xue et al., 2021). Most of the studies focused on natural and processes in the 3800 urban subwatersheds, each characterized by
near-natural watersheds (Senbeta and Romanowicz, 2021), with less unique topography, LULC patterns and drainage networks. Prior to
attention to urban areas. Example such studies on urban watersheds are conducting future scenario analyses, PCSWMM was validated against
Alamdari et al. (2017, 2020) and Hajimohammadi et al. (2018). The historical stream water levels. Future projections of LULC and rainfall
reader is referred to Alamdari and Hogue (2022) and Hathaway et al. were integrated into the hydrologic model. The LULC data for the his­
(2024) for a detailed review of studies about CC impacts on urban torical (2010) and future (2070) scenarios reflected the expected urban
watersheds. development and other land use shifts. Climate data, including rainfall
Additionally, research on projecting the joint impacts of future LULC intensity and frequency, were derived from downscaled global climate
change and CC on watershed hydrology is relatively limited compared to models for various future scenarios. These data provided the basis for
those about the individual impacts, particularly for urban watersheds. simulating how changes in LULC and climate would affect urban runoff.
These limited efforts have focused on a small number of watersheds and To assess the impacts of LULC change and CC on subwatershed hy­
used spatially coarse resolution stormwater models that miss details and drology, six 24-h design rainfall events were selected, representing
spatial heterogeneity in urban watersheds. For instance, Ahmadisharaf recurrence intervals from 5 to 200 years.
et al. (2016, 2021) delineated a 43.7 km2 urban watershed into only Using PCSWMM, future surface runoff for each subwatershed was
eight subwatersheds when modeling watershed hydrology via SWMM, projected under the altered LULC and climate conditions. The model
while Alamdari et al. (2022) studied only 25 subwatersheds in a 197 km2 simulated the hydrologic response of the subwatersheds to each of the
urban watershed. Urban watersheds have several important details (e.g., six design rainfall events for both historical and future scenarios. This
hydraulic infrastructure) and land management practices that call for allowed for the assessment of changes in runoff attributes—peak and
fine-scale models to predict LULC change and CC impacts on hydrology; volume. The final step involved comparative analyses between the his­
previous studies often overlooked such details. Coarse-scale modeling is torical and future scenarios in terms of changes in the two runoff attri­
not adequate for detailed projections of the impacts of LULC change and butes under the individual and combined impacts of LULC change and
CC on urban watersheds. CC. Detailed descriptions of each component of the methodology,
Fine-scale hydrologic models offer a significant advantage in including model parameters, data sources and analytical techniques, are
capturing the spatial heterogeneity and detailed characteristics of urban provided in the subsequent sub-sections. Fig. 1 provides a schematic
watersheds (Cao et al., 2020; Habibi et al., 2019; Wirion et al., 2017). view of the methodological framework.
Such models can more accurately simulate the hydrologic processes,
including the distribution of runoff, infiltration rates and the impact of 2.1. Study area
urban infrastructure on streamflows (Fry and Maxwell, 2015; Widowati,
2017; Zhu et al., 2019). The application of these models becomes crucial The study area included 3800 subwatersheds across Hillsborough
when evaluating the effects of LULC changes and CC scenarios such as County in Midwest Florida, a region characterized by its diverse topo­
altered rainfall patterns and intensities. These stressors collectively in­ graphic and hydrologic features (Fig. 2). The subwatersheds are distinct
fluence the watershed hydrology, necessitating a model capable of in terms of drainage area, surface imperviousness, LULC distribution and
detailed representation of pertinent elements and underlying hydrologic ground slope (Table 1). The drainage area varies from 0.1 to 741.3 ha,
processes. surface imperviousness ranges from 0 to 54% and ground slope vary
This paper presents the application of a fine-scale model, addressing from 0 to 10%. As of 2010, this area was predominantly urbanized, with
the limitations of previous studies that relied on coarser spatial resolu­ more than half of the area classified as urban. Over historical years,
tion, thereby offering a more comprehensive and realistic assessment of urbanization has led to more impervious surfaces (Suttles et al., 2018).
future hydrologic changes in urban watersheds. We evaluated the runoff This increase is a critical factor in altering the watershed hydrology,
response to six 24-h design rainfall events under various LULC and CC leading to reduced infiltration rates and increased surface runoff. The
scenarios. The specific objectives were to: (1) study the individual and study area is projected to undergo further urbanization over the next 80
joint impacts of future LULC change and CC on the runoff attributes; (2) years according to the LULC projections by US Environmental Protection
explore the relationship between changes in the runoff attributes and Agency (USEPA, 2023). This projected change will likely exacerbate the
subwatershed characteristics (drainage area, surface imperviousness hydrologic impacts observed due to increased imperviousness and
and ground slope); and (3) investigate the relationship between the altered LULC patterns.
changes and rainfall recurrence interval. Changes in LULC and CC were The study area has a humid subtropical climate and receives an
extracted from existing projections, while urban hydrologic modeling average annual precipitation of 1321 mm. Several severe rainfall events
was conducted via Personal Computer SWMM (PCSWMM; Computa­ have occurred in the past, including those in November 2014 and June
tional Hydraulics International [CHI], 2023). The innovation of this 2012 with 193.0 mm and 332.7 mm of rainfall over five-day periods.
paper stems from its joint examination of LULC change and CC impacts These events highlight the watershed vulnerability to intense precipi­
on urban watershed hydrology using a fine-scale hydrologic model. This tation, a factor that is likely to be exacerbated by CC. Previous research
approach provides an advanced understanding of hydrologic dynamics on LULC in the Hillsborough Watershed (Fouad, 2009; Hackett, 2011;
in urban watersheds. The application to numerous urban subwatersheds Hurt et al., 2013; Moreno, 2011) provided insights into the effects of
(3,800) near Tampa, Midwest Florida, as well as consideration of diverse LULC changes. However, these studies did not consider the joint impacts
subwatershed characteristics (different drainage areas, ground slopes, of LULC change and CC on the watershed hydrology. Table 1 presents
levels of imperviousness and LULC distributions) and several hydraulic the range and average for characteristics of the study subwatersheds.
structures (e.g., ponds and pipes) enhance the comprehensiveness and
generalizability of our study. The use of fine-scale hydrologic modeling

2
L.A. Mayou et al. Journal of Environmental Management 362 (2024) 121284

Fig. 1. Schematic view of the methodological framework. LULC: Land use/land cover; CC: Climate change; CN: Curve number; PCSWMM: Personal Computer Storm
Water Management Model.

2.2. Climate change imperviousness at 10-year increments through the end of century (year
2100) for the entire United States. Here, we used SSP245 and SSP585
The CC impacts on the runoff attributes were evaluated through the scenarios (hereafter called LULC245 and LULC585) projected via Hadley
impacts on 24-h rainfall event depths for six recurrence intervals: 5-year, Center global environmental model (HadGEM2-ES) Earth system model
10-year, 25-year, 50-year, 100-year and 200-year. This range of recur­ (Collins et al., 2011; Martin et al., 2011). These scenarios represent
rence intervals was chosen to cover a broad spectrum of rainfall events, lower and upper pathways of future greenhouse gas emissions. The
from frequent to rare. For each of these events, the average rainfall ICLUS data has a 30 m spatial resolution for LULC. To derive the future
depth was collected from National Oceanic and Atmospheric Adminis­ surface imperviousness in the study watershed, we used the recom­
tration (NOAA) Atlas 14 (Perrelli and Irvine, 2013). The Tampa Bay mended values by NOAA’s Impervious Surface Analysis Tool (ISAT) for
Airport (NOAA Station ID USW00012842) served as the reference National Land Cover Dataset (NLCD) (NOAA, 2023) and cross-walked
weather station for the study subwatersheds. Natural Resources Con­ the NLCD LULCs to those for ICLUS (Supplementary Table S1). The
servation Service (NRCS) type II was selected as the rainfall temporal surface imperviousness provided inputs for the hydrologic model,
pattern based on US Department of Agriculture (Cronshey, 1986) rec­ PCSWMM. In addition to land surface imperviousness, we accounted for
ommendations for the study area. the future changes in LULC through adjusting the parameters of our
CC impacts on design rainfall events were considered through infiltration method; the details are provided in Section 2.5.
change factors. These factors determine the increase or decrease in
rainfall depth for a given recurrence interval under future climate (year 2.4. Fine-scale hydrologic modeling
2070). To bracket the future changes, we used the 5th and 95th
percentile change factors developed by Irizarry-Ortiz et al. (2022) for Event-based hydrologic modeling for historical and future scenarios
various recurrence intervals across South Florida. The change factors was done via PCSWMM. PCSWMM is an enhanced version of SWMM
were derived based on several global and regional climate models and (Rossman, 2015), with several capabilities such as GIS integration,
multiple future plausible greenhouse gas emission scenarios such as watershed delineation, time series analyses and tools for pre- and
representative climate pathways and shared socioeconomic pathways post-processing of the data. This model has been used in previous studies
(SSPs) as defined by Intergovernmental Panel on Climate Change’s to predict the effects of CC and LULC change in several areas across the
(IPCC’s) Coupled Model Intercomparison Project Phases 5 and 6 (CMIP5 globe (Ahiablame and Shakya, 2016; Dressel, 2014; Perrelli and Irvine,
and 6; O’Neill et al., 2016; Taylor et al., 2012); the percentiles are based 2013; Shaneyfelt et al., 2021; Talbot et al., 2016; Alamdari et al., 2017,
the ensemble models and scenarios evaluated in Irizarry-Ortiz et al. 2020, 2022).
(2022). We applied these change factors to the historical rainfall depths For each 24-h design rainfall event, hydrologic modeling was con­
(Perrelli and Irvine, 2013) to estimate the future rainfall depths for the ducted at a 1-min routing timestep for a two-day time window. Since
six recurrence intervals. The 5th and 95th percentile scenarios are PCSWMM is a semi-distributed model, these data were needed for each
hereafter called CC5 and CC95. This approach generated two distinct of the 3800 subwatersheds, all of which were interconnected in terms of
scenarios representing a range of possible future climate conditions. hydrology. The model was then used to predict the runoff attributes
These adjusted rainfall depths were then input into the PCSWMM to under historical and future conditions for the six design rainfall events.
conduct analyses of both historical and future runoff. By comparing the The simulation options in PCSWMM were set up based on the rec­
model outputs under these scenarios, we assessed how variations in ommendations by the literature (Cronshey, 1986; Rossman, 2015,
rainfall depths, as influenced by CC, would affect surface runoff attri­ 2017). The infiltration method was NRCS’ curve number (CN). The CN
butes in the subwatersheds. values, representing normal antecedent moisture conditions, were
initially derived based on Cronshey (1986) according to the surface
2.3. Land use/land cover change imperviousness and hydrologic soil groups (HSGs) in each sub­
watershed. Dynamic wave was used as the channel routing method to
The historical and future LULC distribution and surface impervi­ represent water movement in open channels and pressured structures.
ousness were acquired from integrated climate and land use scenarios The fine-scale model represents the urban details through including
USEPA (2023). ICLUS provides projected LULC and surface several hydraulic structures that affect water movement and hydrologic

3
L.A. Mayou et al. Journal of Environmental Management 362 (2024) 121284

Fig. 2. Study area and stream gauges.

4
L.A. Mayou et al. Journal of Environmental Management 362 (2024) 121284

Table 1 rainfall time series recorded at Tampa Bay Airport weather station were
Characteristics of the study subwatersheds. collected at the 15-min timestep from NOAA’s National Center for
Watershed characteristic Minimum Maximum Mean Environmental Information (NCEI). These rainfall values were aggre­
gated into an hourly timestep for consistency with our design rainfall
Drainage area (ha) 0.1 741.3 9.8
Ground slope (%) 0.0 10.0 0.7 simulations via PCSWMM.
Surface imperviousness (%) 0.0 54.0 36.5
2.7. Hydrologic modeling scenarios
dynamics across the study watershed. In addition to 3800 sub­
Twenty-four design events, which differed in terms of climate and
watersheds, 5193 storage units (e.g., ponds and lakes), 4187 conduits,
LULC, were modeled under one historical and three future conditions
2516 pipes, 13 pumps, 13 orifices and 6514 weirs were included in the
(Table 3). The corresponding hydrologic models, differed in terms of
model.
rainfall depth, surface imperviousness and CN. There were six models
for historical conditions (year 2010). In these models, we used 2010
2.5. Hydrologic model and data for the study area
imperviousness and calibrated CN values for the six design events. The
remaining 18 scenarios represent future conditions for the same rainfall
To simulate runoff attributes in the subwatersheds, an existing
events, six for CC, six for LULC change and six for joint LULC change and
PCSWMM for Hillsborough Watershed, Florida (CH2M HILL Engineers
CC. When projecting the sole impacts of LULC changes, the other
Inc. 2016) was used. This model was calibrated and validated against
stressor (rainfall) was assumed to be invariant in the future. Similarly,
historical events. We adjusted the PCSWMM in terms of LULC-related
LULC was assumed invariant when analyzing CC. This assumption was
parameters—imperviousness and CN—since the original values were
consistent with previous studies such as (Ahmadisharaf et al., 2020;
uniform across the subwatersheds. Table 2 lists the data used in
Lacher et al., 2019; Schilling et al., 2008).
PCSWMM for modeling historical and future conditions. The HSGs,
which drive CN values, were taken from USDA Web Soil Service (USDA,
2.8. Comparative analyses between historical and future scenarios
2019). These values alongside surface imperviousness were used to es­
timate CN values for each of the 3800 subwatersheds. Additional data,
Analyses were performed to study the relationships between the
including channel properties (e.g., geometry and roughness) and hy­
simulated runoff attributes (peak and volume) and subwatershed char­
draulic infrastructure, were taken from CH2M HILL Engineers Inc.
acteristics. The characteristics of concern were drainage area, impervi­
(CH2M, 2016).
ousness and ground slope. We sought to explore whether certain
subwatersheds experience greater changes in runoff. To do so, we
2.6. Hydrologic model validation
scrutinized the predicted changes in the runoff attributes across the
3800 subwatersheds. We also analyzed changes in the runoff attributes
While the PCSWMM of the study area was calibrated by CH2M HILL
across the six recurrence intervals to investigate whether the changes
Engineers Inc. (CH2M, 2016), we conducted additional evaluations on
are greater in more intense rainfall events. Last, we compared the
the model performance. This was done because we adjusted the
relative changes in the runoff peak and volume for LULC change and CC
PCSWMM in terms of the LULC-related parameters—imperviousness
scenarios to explore which nonstationary stressor (LULC or climate) has
and CN. We compared the simulated maximum stream water levels
the dominant impact in the study subwatersheds.
against historical measurements for six historical events. These events
were chosen based on their intensity and duration as well as the avail­
3. Results and discussion
ability of corresponding observed hydrologic data, ensuring a compre­
hensive assessment of the model performance. Each event represented
3.1. Climate change
distinct conditions in terms of rainfall intensity and duration, providing
a diverse set of tests on the model performance. The performance of
The study findings indicated a trend where changes in rainfall depth
PCSWMM was evaluated in terms of relative error. The established
are more pronounced in higher recurrence intervals (Table 4). This
criteria in hydrologic modeling literature (Ahmadisharaf et al., 2019,
pattern was evident across the range of recurrence intervals. In contrast
2022; Moriasi et al., 2015) was used to judge on the model performance.
to the 5-year event, the projected increase in rainfall depth for the 200-
The historical water levels were taken from eight United States
year event was substantially higher. Historically, the 200-year event had
Geological Survey (USGS) stream gauges. For each historical event,
a rainfall depth of 343 mm. Under future CC scenarios, this event was
projected to rise by 144 mm (41.9% increase).
Table 2
Input data for hydrologic modeling in the study subwatersheds.
3.2. Land use/land cover change
Classification Dataset Spatial Year(s) Source
resolution of data
In 2010, the watershed composition included agriculture, wetland,
Land use/land cover ICLUS 30 m 2010 USEPA (2023)
open water and urban LULCs. However, by 2070, both scenarios pro­
and
2070
jected a shift towards urban land and a decrease in wetlands. Under the
Land surface ICLUS 90 m 2010 USEPA (2023) LULC245 scenario, the urban land increases slightly from 63% to 66%,
imperviousness and while wetlands decrease from 18% to 16%. The LULC585 scenario
2070
Drainage network N/A N/A 2012 CH2M HILL
Engineers Inc. Table 3
(2016) Hydrologic modeling scenarios based on land use/land cover (LULC) and
Rainfall events NOAA N/A 2013 Perrelli and Irvine climate change (CC).
(historical and Atlas 14 (2013)
Modeling scenario Rainfall year LULC year
future) Change N/A 2070 Irizarry-Ortiz et al.
factors (2022) Historical 2010 2010
Hydraulic N/A N/A 2012 CH2M HILL CC 2070 (CC5 or CC95) 2010
infrastructure Engineers Inc. LULC change 2010 2070 (LULC245 or LULC585)
(2016) Joint LULC change and CC 2070 (CC5 or CC95) 2070 (LULC245 or LULC585)

5
L.A. Mayou et al. Journal of Environmental Management 362 (2024) 121284

Table 4
Change factors and 24-h rainfall depths for the six recurrence intervals under climate change (CC).
Recurrence Historical rainfall Change factor for the 5th Change factor for the 95th Future depth (mm) for the 5th Future depth (mm) for the 95th
interval (years) depth (mm) percentile CC scenario percentile CC scenario percentile CC scenario percentile CC scenario

5 133 0.97 1.46 129 194

10 158 0.97 1.56 154 247
25 202 0.95 1.72 192 347
50 242 0.92 1.88 223 456
100 290 0.88 2.09 255 605
200 343 0.85 2.33 292 799

showed a more drastic change, with urban areas expanding from 63% to effects of CC on runoff peak and volume in the 3800 subwatersheds.
72% and wetlands reducing to 12%. These changes differed in the Running each model took ~90 min on a computer with core TM and 16
subwatersheds, in which the maximum change in the developed LULC GB DDR3 RAM.
was 1050% for the SSP245 scenario and 1475% for the SSP585 scenario. To show the importance of detailed representation of urban hetero­
Fig. S1 illustrates these changes in the spatial distribution of LULC for geneity and hydraulic structures in our fine-scale model, we eliminated
the LULC245 scenario. Similarly, Figs. S2–S3 show the changes in the structures and compared the model outputs against those by the
imperviousness and CN values, respectively. Relative changes in surface original PCSWMM. Our comparisons showed that, eliminating the
imperviousness ranged from 0% to 77% across the subwatersheds (in storage units, results in substantial increases in the simulated peak
Scenario SSP585), with an average change of 2.6%. The shift towards streamflow of the channel network (average of ~52% and median of
more urbanized LULCs resulted in increased impervious surfaces and ~99%); the increase was more pronounce during higher recurrence
CNs. The transition to more urbanized areas suggests more runoff gen­ intervals. The increase stems from the role of ponds in attenuating peak
eration, while the loss of natural landscapes like wetlands have profound flows in the stream network. Our comparative analyses showed the
impacts on the watershed hydrology (Koepke, 2017). Wetlands play a importance of ponds in modeling urban runoff; overlooking such
crucial role in water retention and flood mitigation; their loss can lead to structures can lead to substantial biases or potentially forcing the model
increased flood hazards (Acreman and Holden, 2013; Hovis et al., 2021; to unrealistic parameters during the calibration and validation.
Kadykalo and Findlay, 2016).
3.5. Impacts of climate change on surface runoff
3.3. Hydrologic model validation
Fig. 3 shows the relative changes between the historical (year 2010)
The validation of the PCSWMM was conducted against six historical and future (under CC for year 2070) values. For both runoff peak and
rainfall events. For each historical event, the model was run using hourly volume, the relative change was greater in the higher recurrence in­
rainfall time series from the representative weather station tervals. That is, CC-induced changes in runoff were greater in rainfall
(USW00012842). Each model was run on a computer equipped with a events with larger depths. The median of the relative changes (across the
Core™ i7-11700 2.5 GHz CPU and 16 GB DDR3 RAM, with each simu­ 3800 subwatersheds) was ~25% and ~60% for the 5-year and 200-year
lation taking approximately 3 h. events, respectively.
The predicted maximum water levels by the PCSWMM were derived We compared the changes in runoff against historical imperviousness
at the nodes that correspond to the eight USGS stream gauges. These of the 3800 subwatersheds to evaluate whether subwatersheds that are
predicted water levels were then compared with the observed data at the more impervious experience a greater change in the future. The results
gauges (Table 5). The comparisons revealed that the average relative showed that, only in the smallest recurrence interval (5-year) and for
error of the predicted water levels ranged from − 4% to 18%. This range runoff volume, greater changes were generally evident in subwatersheds
was considered within acceptable limits according to the criteria by with higher imperviousness. This was inconsistent with Caldwell et al.
Ahmadisharaf et al. (2019, 2022) and Moriasi et al. (2015), particularly (2012) who found that increases in streamflow are evident in areas that
given the complexities involved in our urban watershed. The model are readily developed. Such results were not evident in greater recur­
outputs were also found to be comparable with previous studies (CH2M, rence intervals.
2016), further reinforcing the PCSWMM efficiency. This validated The results of the six PCSWMM models were also compared against
model was used for the subsequent analyses of CC and LULC scenarios. the drainage area of the subwatersheds. Our results showed that, overall,
the smaller the subwatershed, the greater the change in runoff volume
and peak. In addition, smaller subwatersheds exhibiting greater changes
3.4. Fine-scale hydrologic modeling of design events
in runoff volume and peak indicated a scale-dependency in hydrologic
responses to CC. These smaller subwatersheds might lack the natural
For each modeling scenario, either historical or future, six PCSWMM
storage and saturation, making them more sensitive to rapid changes in
models (one for each recurrence interval) were run to simulate the
water volume due to intense rainfall events.
The results in terms of ground slope also indicated that, in general,
Table 5 the milder the watershed, the higher the relative change in the runoff
Modeled and observed water levels during the validation events. USGS: US
attributes. This pattern was more prominent in greater recurrence in­
Geological Survey.
tervals. The greater changes in runoff in the higher recurrence intervals
Event date USGS stream Modeled water Observed water suggested that more intense rainfall events, which are typical under
gauge ID level (m) level (m)
many CC scenarios, lead to significantly higher runoff. This can be due to
6/24/2012–6/28/ 2,306,647 4.62 3.85 the inability of the ground and existing drainage systems to absorb or
2012 2,306,904 13.39 No data
retain sudden large volumes of water effectively.
2,306,774 8.54 7.81
2,307,323 8.72 9.09 The absence of a consistent relationship across all recurrence in­
11/24/2014–11/ 2,306,647 4.38 3.60 tervals suggested that factors beyond imperviousness, such as rainfall
28/2014 2,306,904 13.33 12.42 intensity and land use practices, play a more significant role in influ­
2,306,774 8.38 6.71 encing runoff changes under CC. This challenges traditional assumptions
2,307,323 8.71 8.69
linking impervious surfaces directly to increased runoff.

6
L.A. Mayou et al. Journal of Environmental Management 362 (2024) 121284

Fig. 3. Predicted changes under joint impacts of climate change (CC) in the future (year 2070) for: (a) 5th percentile climate (runoff volume); (b) 95th percentile
climate SSP245 LULC (runoff volume); (c) 5th percentile climate (peak runoff); and (d) 95th percentile climate (peak runoff) for six design rainfall events.

3.6. Impacts of land use/land cover change on surface runoff subwatersheds that were generally larger than this study.
Similar to the results for the runoff volume, there was a negative
The relative change in runoff volume (from 2010 to 2070) was relationship between the relative change in peak runoff and drainage
compared to the surface imperviousness of the subwatersheds under area. Our results showed that the smaller the subwatershed, the greater
both the SSP245 and SSP585 scenarios (Fig. 4). There was no relation­ the relative change. For the 5-year recurrence interval, subwatersheds
ship between the imperviousness and the relative change in runoff with a drainage area <40 ha had relative changes <350%, while sub­
volume. Similarly, there was no relationship between the relative watersheds with areas >40 ha had relative changes of near zero. The
change in peak runoff with respect to the recurrence interval; the rela­ more severe the rainfall, the smaller the relative change. For the 5-year
tive changes across the 3800 subwatersheds varied between 0% and recurrence interval, the relative change values were mostly between 0%
350% for 5-year and 200-year events, respectively. There was also no and 350%, while for the 200-year event, the values were mostly between
relationship between the increase in imperviousness and the relative 0% and 250%.
change in peak runoff. The relative changes were also compared to the subwatershed slope.
A negative relationship was generally evident between the relative The runoff volume results did not indicate any relationships between the
change in peak runoff and subwatershed drainage area; smaller sub­ relative change and the rainfall recurrence interval; all of the relative
watersheds exhibited greater changes in peak runoff. This was more changes remain <120% for the six recurrence intervals. Similar to the
evident in higher recurrence intervals. For instance, in the 200-year findings for CC, the subwatersheds with milder slopes had a larger
event, a subwatershed with a drainage area of 2 ha had ~120% relative change in runoff volume. The peak runoff, however, showed
change in runoff volume, while a ~80 ha subwatershed had a relative higher relative changes for the lower recurrence intervals. The 5-year
change of ~20%. While the relationship strengthened with the increase event led to a maximum relative change of 350%, while the 200-year
in rainfall recurrence intervals, the overall relative change stayed about rainfall yielded a maximum change of 250%. The subwatersheds with
the same range for all of the rainfall events (between 0% and 140%). Our milder slopes still produced larger relative changes in terms of peak
findings about the greater impacts of LULC change on runoff volume in runoff.
smaller subwatersheds are in agreement with previous studies (e.g., The lack of correlation between subwatershed imperviousness and
Ahmadisharaf et al., 2020; Lacher et al., 2019) for mixed land use runoff volume change suggested that impervious surfaces may not be the

7
L.A. Mayou et al. Journal of Environmental Management 362 (2024) 121284

Fig. 4. Predicted changes under future (year 2070) land use/land cover for: (a) SSP245 (runoff volume); (b) SSP585 (runoff volume); (c) SSP245 (peak runoff); and
(d) SSP585 (peak runoff) for six design rainfall events.

primary driver of runoff in these subwatersheds. This can be due to 3.7. Joint impacts of climate and land use/land cover change on surface
factors like the presence of effective drainage systems, soil types that runoff
facilitate better water infiltration or vegetation that absorbs and slows
down runoff. The absence of a relationship between peak runoff and The same six recurrence intervals were also used to study the joint
both imperviousness and recurrence interval indicated that peak runoff impacts of CC and LULC change. Fig. 5 shows the relative change in the
responses may be more influenced by other hydrologic factors, such as peak runoff across the 3800 subwatersheds. Most subwatersheds had a
subwatershed shape, soil saturation levels or the intensity and duration relative change of <0.5% in runoff volume. There were a limited num­
of rainfall events, rather than surface imperviousness or the recurrence ber of subwatersheds scattered across the center of the subwatersheds
interval. The finding that subwatersheds with different levels of with greater relative changes. The relative change in the subwatersheds
imperviousness show similar relative changes in runoff for the 200-year was also predicted to be higher in greater recurrence intervals.
rainfall event suggested that there may be a threshold of imperviousness The relative change in runoff volume was compared to the imper­
beyond which its impact on runoff change is not significant. This can be viousness of the subwatersheds. The results did not indicate that there
due to compensatory mechanisms in the subwatersheds, such as are any relationships between these two variables. For the 5-year event,
enhanced drainage or storage capacity. The greater changes in peak the average relative change in runoff volume was greatest among the
runoff in smaller subwatersheds can be attributed to the limited space recurrence intervals. As the recurrence interval increases, the relative
for water storage and slower hydrologic processes. In smaller sub­ changes in runoff volume became smaller; the results for the peak runoff
watersheds, surface runoff has less area to spread, leading to quicker and demonstrated a similar relationship to the runoff volume results. Similar
more significant runoff responses. The strengthening of this relationship to the volume, our results showed no relationship between peak runoff
with higher recurrence intervals can be because smaller subwatersheds and subwatershed imperviousness. For the 5-year event, the average
reaching their capacity to handle water more quickly during severe relative change was ~100% for the subwatersheds with an impervi­
rainfall events. The observation of larger relative changes in runoff for ousness of 10% and 50%. The relative changes in peak runoff for the 5-
milder subwatersheds may be attributed to the fact that these sub­ year event were mostly between 0% and 200%, while for the 200-year
watersheds retain more water, leading to higher relative changes in event, the values were mostly between 50% and 150%.
runoff, especially during heavy rainfalls. The analyses showed that smaller subwatersheds experience greater

8
L.A. Mayou et al. Journal of Environmental Management 362 (2024) 121284

Fig. 5. Predicted changes under joint impacts of climate change (CC) and land use/land cover (LULC) change in the future (year 2070) for scenarios: (a) 5th
percentile climate and LULC SSP245 (runoff volume); (b) 5th percentile climate and LULC SSP585 (runoff volume); (c) 95th percentile climate and LULC SSP245
(runoff volume); (d) 95th percentile climate and LULC SSP585 (runoff volume); (e) 5th percentile climate and LULC SSP245 (peak runoff); (f) 5th percentile climate
and LULC SSP585 (peak runoff); (g) 95th percentile climate and LULC SSP245 (peak runoff); and (h) 95th percentile climate and LULC SSP585 (peak runoff) for six
design rainfall events.

9
L.A. Mayou et al. Journal of Environmental Management 362 (2024) 121284

relative changes in runoff volume. For the 5-year event, subwatersheds strong (<0.41). Most of the correlations were also not statistically sig­
with an area <40.5 ha had relative changes up to ~400%, while all nificant (p-value>0.05). Therefore, the statistical significance of our
subwatersheds with areas over 80.9 ha had a relative change of near findings about the relationships between the watershed characteristics
zero. The greater the recurrence interval, the smaller the relative and changes in the runoff attributes should be considered when using
changes. For the 5-year event, the relative changes were mostly between our study results.
0% and 200%, while for the 200-year event, the values were mostly
between 30% and 80%. Similar to the results for the runoff volume, the 3.8. Comparative analyses among the future scenarios
results for the peak runoff had a negative relationship between the peak
runoff and the relative change. The relative change in peak runoff was Fig. 6 compares the results of the eight modeling scenarios for each
compared to the drainage area of the subwatersheds. Smaller sub­ rainfall event regarding the two runoff attributes. Overall, the peak
watersheds had higher relative changes. For the 5-year event, sub­ runoff values had greater variability and more outliers than the runoff
watersheds with an area <40.5 ha had relative changes of up to ~400%, volume. Of the two nonstationary stressors, CC had the greater impact.
while all the subwatersheds with areas >80.9 ha had relative changes of The CC95 scenario resulted in greater changes in terms of both runoff
near zero. attributes. Previous research (Chawla and Mujumdar, 2015; Guo et al.,
Similar to the result for the LULC change, the more severe the rain­ 2020; Gupta et al., 2015) found inconsistent results about the domi­
fall, the lower the relative change and the less variability among the nance of CC over LULC change. The inconsistencies stem from the study
relative changes in the 3800 subwatersheds. For the 5-year event, the area characteristics and assumptions about the future scenarios. While
relative changes varied between 0% and 250%, while for the 200-year such comparative studies are rare for urban areas, our findings about the
event, the changes were between 50% and 200%. Our analyses also dominance of CC over LULC change, were inconsistent with what was
indicated that subwatersheds with steeper slopes produced higher found by Alamdari et al. (2022) for a rapidly growing urban sub­
relative changes in runoff volume. watershed in Virginia. This inconsistency is most likely due the nature of
Previous studies (e.g., Chen et al., 2017; Dhakal and Chevalier, 2017; the study area and modeling scenarios. The future scenarios in Alamdari
Du et al., 2012; Fletcher et al., 2015; Jennings and Taylor Jarnagin, et al. (2022) showed relatively greater changes in LULC compared to us.
2002; Meerow and Newell, 2021; Zhou, 2014) provided evidence of how Conversely, our CC scenarios showed more aggressive changes in rain­
even minor increases in imperviousness can lead to major changes in fall depth than Alamdari et al. (2022). In watersheds with more
runoff peak and volume. The magnitude of this increase varies, but is a aggressive CC projections or lesser LULC changes, the impact of CC on
consistent indicator of urban hydrologic change. The impact of imper­ runoff would naturally be more pronounced.
vious surfaces on urban hydrology is not solely a function of its quantity As the recurrence interval increases, the effects of CC increase and
but also its spatial distribution and connectivity. Impervious surfaces, the effects of LULC decrease. For the 200-yr rainfall, changes under the
which are highly connected and located near drainage networks, can CC and joint CC-LULC change scenarios are almost equal, while the
significantly amplify peak flow rates (Dhakal and Chevalier, 2017; LULC is almost equal to the historical scenario; there was not a signifi­
Fletcher et al., 2015). This was not evident in our study though. The cant difference between the LULC245 and LULC585 scenarios.
absence of a relationship between subwatershed imperviousness and The results discussed above indicate that there is a larger change in
runoff volume changes indicated that other factors, such as microcli­ runoff volume and peak for greater recurrence intervals. This was ex­
mate conditions, land management practices or specific geomorpho­ pected as higher change factors were projected for larger recurrence
logical features, soil permeability, vegetation or local topography, may intervals by Irizarry-Ortiz et al. (2022), which was used in this research.
play a more important role in influencing runoff. The consistent relative It is also consistent with general consensus that CC affects extreme
change in runoff volume across subwatersheds with different levels of events more severely, leading to more severe runoff conditions (Arnone
imperviousness suggested a potential threshold effect, where impervi­ et al., 2018; Herrador et al., n.d.; Khan et al., 2015).
ousness impacts runoff up to a certain point, beyond which other factors The historical imperviousness of the subwatersheds showed no re­
become more dominant. Similar to runoff volume, the lack of a clear lationships with the relative change in the runoff attributes. Both runoff
correlation between peak runoff and subwatershed imperviousness volume and peak had a greater increase in smaller subwatersheds. This
might be explained by the overriding influence of other subwatershed was align with Hurt et al. (2013) who found that smaller subwatershed
characteristics such as soil type, slope or drainage efficiency. areas are more susceptible to higher runoff peaks and volumes. Overall,
The decreasing variability in peak runoff with increasing recurrence the higher the recurrence interval, the more variability in the changes.
intervals indicated a saturation effect where, beyond a certain intensity The changes were also generally more pronounced in steeper
of rainfall, the impact of impervious surfaces on peak runoff becomes subwatersheds.
less pronounced. The higher relative changes in runoff observed in Even though CC was found as the dominant nonstationary stressor
smaller subwatersheds can be due to a lack of sufficient area to absorb for runoff changes, LULC changes still contributed to runoff variations,
and dissipate rainfall, leading to more rapid runoff. Larger sub­ especially the subwatersheds identified as highly sensitive to LULC
watersheds likely have more varied LULC and topography, which can changes (e., smaller subwatersheds). These results were in line with the
help disperse and absorb rainfall, thereby reducing the relative change analyses of Fouad (2009) and Xian (2007) who indicated that even small
in runoff. The observation that more severe rainfall leads to lower changes in natural land area significantly impact the local watershed
relative changes and less variability among the subwatersheds might be hydrology. Our results suggested that LULC-induced hydrologic re­
due to the saturation of the landscape, which becomes more effective sponses are highly localized and must be considered in conjunction with
under extreme weather conditions (e.g., 200-year event). Sub­ CC for effective land use planning and proposing management strategies
watersheds with steeper slopes leading to greater relative changes in that mitigate potential adverse effects on runoff.
runoff volume suggest that slope gradient significantly impacts flow Our results showed the need for enhancing the capacity of current
velocity and, consequently, runoff generation. Steeper slopes may hydraulic infrastructure systems, a need for regulations updates, land
accelerate surface runoff, leading to higher relative changes in runoff use zoning and mitigation strategies accordingly. This paper is relevant
volume. for urban planners and floodplain managers that aim to maintain the
In addition to the visual comparisons on the relationship between the functionality of drainage networks. Building climate resilience in cities
relative changes in runoff attributes and watershed characteristics, we requires research like this paper to guide development of effective ac­
conducted correlation analyses using Pearson’s r (linear) and Kendall’s τ tions and strategies while reducing CC-induced changes in runoff. Co­
(rank-based). These analyses show that the correlations between relative ordination beyond only one urban subwatershed and effective land use
changes in the runoff attributes and watershed characteristics are not planning at the city scale can balance these changes out and avoid

10
L.A. Mayou et al. Journal of Environmental Management 362 (2024) 121284

Fig. 6. Boxplots of changes in runoff: (a) volume; and (b) peak across the 3800 subwatersheds for 5-year, 10-year, 25-year, 50-year, 100-year and 200-year events
under historical, climate change (CC), land use/land cover (LULC) change and joint CC and LULC change (CC&LULC).

abnormal runoff conditions and flooding. To prevent abnormal runoff conditions and mitigate flood risks,
The findings underscore the importance of updating regulations and cities need to adopt an integrated water resource management
land use zoning to accommodate the changing hydrologic realities. This approach. This includes not only managing the quantity of runoff but
includes revising building codes, implementing stricter controls on land also improving the quality of urban runoff by reducing pollution and
development in flood-prone areas and ensuring that new developments enhancing natural water cycles. The study’s findings should inform
integrate sustainable drainage systems. There is a need for compre­ policy decisions related to urban development and environmental sus­
hensive mitigation strategies that address the increased flood risks. tainability. Policymakers need to consider the long-term implications of
These strategies could involve a combination of structural measures CC and LULC changes on urban water systems and integrate these
(like flood barriers) and non-structural measures (e.g., public awareness considerations into environmental policies and urban development
campaigns and emergency preparedness plans). The research provides plans. While the study focuses on a specific region, its findings have
valuable insights for urban planners and floodplain managers in devel­ global relevance. Urban areas worldwide facing similar CC and urban­
oping strategies to maintain the functionality of drainage networks. It ization challenges can learn from these insights. The methodology and
emphasizes the need for planning that anticipates future climatic con­ approach used in this study could be adapted to other urban settings,
ditions and integrates resilience into urban design. Building climate providing a framework for assessing and addressing CC impacts on
resilience is critical for cities facing increased CC impacts. This requires urban hydrology.
a holistic approach that combines infrastructure development, com­
munity engagement and policy reforms to enhance the city’s capacity to
withstand and adapt to changing weather patterns. Effective manage­ 3.9. Limitations and future research directions
ment of runoff and flood risks requires coordination beyond individual
urban watersheds. A city-scale approach to land use planning that We only studied on one urban watershed. To find more inclusive
considers the cumulative impact of individual subwatersheds is essen­ results, multiple urban watersheds with diverse characteristics would be
tial. This could involve regional collaboration among municipalities, necessary. The study watershed has a subtropical climate and overall
particularly in areas sharing subwatersheds or drainage systems. mild ground slope. CC impacts were solely considered in rainfall depth,
while it is expected that other rainfall characteristics, such as temporal

11
L.A. Mayou et al. Journal of Environmental Management 362 (2024) 121284

Fig. 6. (continued).

pattern, are affected; these can affect the runoff attributes. LULC impacts 369.9 km2 around Hillsborough County near Tampa, Florida. These
were mainly represented by changing the surface imperviousness. subwatersheds were distinct in terms of drainage area, ground slope,
However, such impacts can result in changes in land management imperviousness and LULC distribution. PCSWMM was used for hydro­
practices (e.g., green infrastructure) and drainage network; these were logic modeling. The model was validated against multiple historical
not incorporated in this research. While the PCSWMM was validated events. The validated model was then applied to simulate the runoff
against historical events, it is subject to uncertainty arising from model attributes for historical and future conditions, representing years 2010
structure and parameterization (Ahmadisharaf et al., 2019, 2022; Moges (historical) and 2070 (future). For CC, change factors were applied to
et al., 2021; Sun et al., 2014). In projecting future changes, we applied historical design rainfall depths. Existing projections (ICLUS dataset)
PCSWMM deterministically. Future research should study the use of were used to represent future LULC. CN values and surface impervi­
probabilistic models and address such intra-model uncertainties. How­ ousness were also adjusted based on the latter for the future conditions.
ever, probabilistic modeling of the study watershed via the fine-scale For each of the six recurrence intervals, runoff predictions were
model would require substantial computational expenses. This makes made for one historical and three future scenarios—CC, LULC change
the probabilistic applications impractical with the current typical and subjoint CC-LULC change—in the study 3800 subwatershed. Over­
computing resources. all, we found that changes in peak runoff were greater than volume. The
results indicated that CC and LULC change are inversely related to the
4. Summary and conclusions characteristics of subwatersheds (drainage area and ground slope) and
directly related to the design rainfall recurrence interval. Such re­
This paper investigated the effects of CC and LULC changes on two lationships were not found between runoff and subwatershed impervi­
runoff attributes—volume and peak—in an urban watershed. The ge­ ousness. However, the findings about the relationship between the
ography of focus included 3800 subwatersheds spanning a total area of watershed characteristics and changes in the runoff attributes was not

12
L.A. Mayou et al. Journal of Environmental Management 362 (2024) 121284

statistically significant; care should be taken when using these findings References
in future studies. The main findings can be summarized as.
Acreman, M., Holden, J., 2013. How wetlands affect floods. Wetlands 33, 773–786.
Ahiablame, L., Shakya, R., 2016. Modeling flood reduction effects of low impact
• Future changes in peak runoff were predicted to be larger than development at a watershed scale. J. Environ. Manag. 171, 81–91.
volume. Ahmadisharaf, E., Alamdari, N., Tajrishy, M., Ghanbari, S., 2021. Effectiveness of
• The relative change in runoff values was greater under more severe retention ponds for sustainable urban flood mitigation across range of storm depths
in northern Tehran, Iran. J. Sustain. Water in the Built Environ. 7 (2), 05021003.
rainfall events in the future (higher recurrence intervals). Ahmadisharaf, E., Camacho, R.A., Zhang, H.X., Hantush, M.M., Mohamoud, Y.M., 2019.
• CC had a greater impact on the runoff attributes than LULC change. Calibration and validation of watershed models and advances in uncertainty analysis
One reason for this was that, in the study subwatershed, the changes in TMDL studies. J. Hydrol. Eng. 24 (7), 03119001.
Ahmadisharaf, E., Camacho-Rincon, R.A., Hewitt, H.X., Hantush, M., Mohamoud, Y.,
in LULC were much smaller than rainfall depth. That is, care should 2022. Model calibration and validation. In: Zhang, H.X., Quinn, N.W.T., Borah, D.K.,
be taken to use this finding in other watersheds. Padmanabhan, G. (Eds.), TMDL Manual of Practice. ASCE, Arlington, VA.
• Smaller subwatersheds experienced higher relative changes in the Ahmadisharaf, E., Lacher, I.L., Fergus, C., Benham, B.L., Akre, T., Kline, K.S., 2020.
Projecting land use change impacts on nutrients, sediment and runoff in multiple
runoff attributes.
spatial scales: business-as-usual vs. Stakeholder-informed scenarios. J. Clean. Prod.
• Subwatersheds with milder ground slope had greater relative 257, 120466.
changes in the runoff attributes. Ahmadisharaf, E., Tajrishy, M., Alamdari, N., 2016. Integrating flood hazard into site
• The historical subwatershed imperviousness did not exhibit any re­ selection of detention basins using spatial multi-criteria decision-making. J. Environ.
Plann. Manag. 59 (8), 1397–1417.
lationships with the relative change in runoff. Alamdari, N., Claggett, P., Sample, D.J., Easton, Z.M., Yazdi, M.N., 2022. Evaluating the
joint effects of climate and land use change on runoff and pollutant loading in a
The findings of this paper help urban planners and floodplain man­ rapidly developing watershed. J. Clean. Prod. 330, 129953.
Alamdari, N., Hogue, T.S., 2022. Assessing the effects of climate change on urban
agers identify the required actions and strategies to protect urban wa­ watersheds: a review and call for future research. Environ. Rev. 30 (1), 61–71.
tersheds against future changes. Our results about LULC change impacts Alamdari, N., Sample, D.J., Ross, A.C., Easton, Z.M., 2020. Evaluating the impact of
can guide zoning strategies from a surface runoff management climate change on water quality and quantity in an urban watershed using an
ensemble approach. Estuar. Coast 43, 56–72.
perspective, while the findings related to CC impacts can help planning Alamdari, N., Sample, D.J., Steinberg, P., Ross, A.C., Easton, Z.M., 2017. Assessing the
for strategies that enhance the resilience of our communities and civil effects of climate change on water quantity and quality in an urban watershed using
infrastructure. The findings from this research contribute to the devel­ a calibrated stormwater model. Water 9 (7), 464.
Arnbjerg-Nielsen, K., Willems, P., Olsson, J., Beecham, S., Pathirana, A., Bülow
opment of more sustainable and resilient urban water management Gregersen, I., Madsen, H., Nguyen, V.-T.-V., 2013. Impacts of climate change on
practices. This study advances the scientific understanding of urban rainfall extremes and urban drainage systems: a review. Water Sci. Technol. 68 (1),
watershed dynamics under nonstationary conditions and serves as a 16–28.
Arnone, E., Pumo, D., Francipane, A., Loggia, G., Noto, L., 2018. The role of urban
vital resource for policymakers and stakeholders involved in urban
growth, climate change, and their interplay in altering runoff extremes. Hydrol.
development and environmental conservation. Process. 32, 1755–1770.
Caldwell, P.W., Sun, G., McNulty, S.G., Cohen, E.C., Moore Myers, J.A., 2012. Impacts of
CRediT authorship contribution statement impervious cover, water withdrawals, and climate change on river flows in the
conterminous US. Hydrol. Earth Syst. Sci. 16 (8), 2839–2857.
Cao, X., Lyu, H., Ni, G., Tian, F., Ma, Y., Grimmond, C., 2020. Spatial scale effect of
Lauren Ashley Mayou: Writing – original draft, Visualization, surface routing and its parameter upscaling for urban flood simulation using a grid-
Validation, Formal analysis, Data curation. Nasrin Alamdari: Writing – based model. Water Resour. Res. 56.
CH2M, 2016. Hillsborough County Northwest Five Watershed Management Plan Update.
review & editing, Supervision, Methodology. Ebrahim Ahmadisharaf: CH2M HILL Engineers, Inc.
Writing – review & editing, Supervision, Methodology, Conceptualiza­ Chawla, I., Mujumdar, P.P., 2015. Isolating the impacts of land use and climate change
tion. Meysam Kamali: Visualization, Formal analysis, Data curation. on streamflow. Hydrol. Earth Syst. Sci. 19 (8), 3633–3651.
Chen, J., Theller, L., Gitau, M.W., Engel, B.A., Harbor, J.M., 2017. Urbanization impacts
on surface runoff of the contiguous United States. J. Environ. Manag. 187, 470–481.
Declaration of competing interest Chen, Q., Chen, H., Wang, J., Zhao, Y., Chen, J., Xu, C., 2019. Impacts of climate change
and land-use change on hydrological extremes in the Jinsha River Basin. Water 11
(7), 1398.
The authors declare the following financial interests/personal re­ CHI, 2023. Personnel Computer Storm Water Management Model by Computational
lationships which may be considered as potential competing interests: Hydraulics International. .
Nasrin Alamdari reports financial support was provided by Florida State Collins, W.J., Bellouin, N., Doutriaux-Boucher, M., Gedney, N., Halloran, P., Hinton, T.,
Hughes, J., Jones, C.D., Joshi, M., Liddicoat, S., Martin, G., O’Connor, F., Rae, J.,
University. Nasrin Alamdari reports a relationship with Florida State
Senior, C., Sitch, S., Totterdell, I., Wiltshire, A., Woodward, S., 2011. Development
University that includes: employment and funding grants. and evaluation of an earth-system model HadGEM2, geosci. Model Dev. Discuss. 4,
997–1062.
Data availability Cronshey, R., 1986. Urban Hydrology for Small Watersheds (Issue 55). US Department of
Agriculture, Soil Conservation Service, Engineering Division.
Dhakal, K.P., Chevalier, L.R., 2017. Managing urban stormwater for urban sustainability:
Data will be made available on request. barriers and policy solutions for green infrastructure application. J. Environ. Manag.
203, 171–181.
Dressel, L., 2014. The Impact of Rochester Storm Sewers on the Water Quality of the
Acknowledgements Lower Genesee River: A Modeling Approach Using PCSWMM.
Du, J., Qian, L., Rui, H., Zuo, T., Zheng, D., Xu, Y., Xu, C.-Y., 2012. Assessing the effects
We thank CHI for providing academic license of PCSWMM to Lauren of urbanization on annual runoff and flood events using an integrated hydrological
modeling system for Qinhuai River basin, China. J. Hydrol. 464, 127–139.
A. Mayou and Hillsborough County for providing the SWMM files for the Fletcher, T.D., Shuster, W., Hunt, W.F., Ashley, R., Butler, D., Arthur, S., Trowsdale, S.,
study area. Nasrin Alamdari was partially supported through a grant by Barraud, S., Semadeni-Davies, A., Bertrand-Krajewski, J.-L., 2015. SUDS, LID, BMPs,
Tampa Bay Estuary Program’s Tampa Bay Environmental Restoration WSUD and more–The evolution and application of terminology surrounding urban
drainage. Urban Water J. 12 (7), 525–542.
Fund (https://tbep.org/). Fouad, G.G., 2009. Assessing the Performance of Water Bodies in Hillsborough County,
Florida Using Data Envelopment Analysis (DEA).
Appendix A. Supplementary data Fowler, H.J., Ali, H., Allan, R.P., Ban, N., Barbero, R., Berg, P., Blenkinsop, S., Cabi, N.S.,
Chan, S., Dale, M., 2021. Towards advancing scientific knowledge of climate change
impacts on short-duration rainfall extremes. Philosoph. Transac. Royal Soc. A 379
Supplementary data to this article can be found online at https://doi. (2195), 20190542.
org/10.1016/j.jenvman.2024.121284. Fry, T., Maxwell, R., 2015. Evaluation of distributed BMPs in an urban watershed—high
resolution modeling for stormwater management. Hydrol. Process. 31, 2700–2712.
Guo, Y., Fang, G., Xu, Y.P., Tian, X., Xie, J., 2020. Identifying how future climate and
land use/cover changes impact streamflow in Xinanjiang Basin, East China. Sci.
Total Environ. 710, 136275.

13
L.A. Mayou et al. Journal of Environmental Management 362 (2024) 121284

Gupta, S.C., Kessler, A.C., Brown, M.K., Zvomuya, F., 2015. Climate and agricultural land Ndulue, E., Mbajiorgu, C., Ugwu, S., Ogwo, V., Ogbu, K., 2015. Assessment of land use/
use change impacts on streamflow in the upper midwestern United States. Water cover impacts on runoff and sediment yield using hydrologic models: a review.
Resour. Res. 51 (7), 5301–5317. J. Ecol. Nat. Environ. 7, 46–55.
Habibi, H., Dasgupta, I., Noh, S., Kim, S., Zink, M., Seo, D., Bartos, M., Kerkez, B., 2019. Neitsch, S.L., Arnold, J.G., Kiniry, J.R., Williams, J.R., 2011. Soil and Water Assessment
High-resolution hydrologic forecasting for very large urban areas. J 21 (3), 441–454. Tool Theoretical Documentation Version 2009. Texas Water Resources Institute.
Hackett, K.E., 2011. Statistical Analysis and Mechanistic Modeling of Water Quality: Ni, X., Parajuli, P.B., Ouyang, Y., Dash, P., Siegert, C., 2021. Assessing land use change
Hillsborough Bay. University of South Florida, Florida. impact on stream discharge and stream water quality in an agricultural watershed.
Hajimohammadi, M., Azizian, A., Ghermezcheshmeh, B., 2018. Evaluation of the impact Catena 198, 105055.
of climate change on runoff in Kan Watershed. Water Eng. Manag. 10 (2), 144–156. NOAA, 2023. Impervious surface analysis tool. https://coast.noaa.gov/digitalcoast/too
Hansen, A.T., Campbell, T., Cho, S.J., Czuba, J.A., Dalzell, B.J., Dolph, C.L., ls/isat.html.
Hawthorne, P.L., Rabotyagov, S., Lang, Z., Kumarasamy, K., 2021. Integrated O’Neill, B.C., Tebaldi, C., Van Vuuren, D.P., Eyring, V., Friedlingstein, P., Hurtt, G.,
assessment modeling reveals near-channel management as cost-effective to improve Knutti, R., Kriegler, E., Lamarque, J.-F., Lowe, J., 2016. The scenario model
water quality in agricultural watersheds. Proc. Natl. Acad. Sci. USA 118 (28), intercomparison project (ScenarioMIP) for CMIP6. Geosci. Model Dev. (GMD) 9 (9),
e2024912118. 3461–3482.
Hathaway, J.M., Bean, E.Z., Bernagros, J.T., Christian, D.P., Davani, H., Ebrahimian, A., Perrelli, M., Irvine, K., 2013. Planning level modeling of E. coli levels in a suburban
Fairbaugh, C.M., Gulliver, J.S., McPhillips, L.E., Palino, G., 2024. A synthesis of watershed using PCSWMM. Pragmatic Model. Urban Water Syst. 21, 423–435.
climate change impacts on stormwater management systems: designing for resiliency Monograph.
and future challenges. J. Sustain. Water in the Built Environ. 10 (2), 04023014. Rajaei, F., Dahmardeh Behrooz, R., Ahmadisharaf, E., Galalizadeh, S., Dudic, B.,
Herrador, B., Lund, V., Fonahn, W., Hisdal, H., Hygen, H., Hyllestad, S., Nordeng, Z., Spalevic, V., Novicevic, R., 2021. Application of integrated watershed management
Skaland, R., Sunde, L., Vold, L., White, R., Wong, W., & Nygård, K. (n.d.). Heavy measures to minimize the land use change impacts. Water 13 (15), 2039.
weather events, water quality and gastroenteritis in Norway. One Health, 13. https Rossman, L., 2015. Storm water management model reference manual: Volume I – Hydrology.
://doi.org/10.1016/j.onehlt.2021.100297. Natl. Risk Manag. Lab., Off. Res. Dev. USEPA. Report No.: EPA/600/R-15/162A. htt
Hovenga, P.A., Wang, D., Medeiros, S.C., Hagen, S.C., Alizad, K., 2016. The response of ps://nepis.epa.gov/Exe/ZyPDF.cgi?Dockey=P100NYRA.txt.
runoff and sediment loading in the Apalachicola River, Florida to climate and land Rossman, L., 2017. Storm Water Management Model reference manual: Volume II –
use land cover change. Earth’s Future 4 (5), 124–142. Hydraulics. Natl. Risk Manag. Lab., Off. Res. Dev., USEPA. Report No. EPA/600/R-
Hovis, M., Hollinger, J., Cubbage, F., Shear, T., Doll, B., Kurki-Fox, J., Line, D., Fox, A., 17/111. https://nepis.epa.gov/Exe/ZyPDF.cgi?Dockey=P100S9AS.pdf.
Baldwin, M., Klondike, T., Lovejoy, M., Evans, B., West, J., Potter, T., 2021. Natural Sadeghi, S., Saghafian, B., Najarchi, M., 2020. Assessment of impacts of change in land
infrastructure practices as potential flood storage and reduction for farms and rural use and climatic variables on runoff in Tajan River Basin. Water Supply 20 (7),
communities in the North Carolina coastal plain. Sustainability. 2779–2793.
Huq, E., Abdul-Aziz, O.I., 2021. Climate and land cover change impacts on stormwater Schilling, K.E., Jha, M.K., Zhang, Y., Gassman, P.W., Wolter, C.F., 2008. Impact of land
runoff in large-scale coastal-urban environments. Sci. Total Environ. 778, 146017. use and land cover change on the water balance of a large agricultural watershed:
Hurt, G.W., Mylavarapu, R.S., Boetger, S.P., 2013. UF/IFAS Nutrient Management Series: historical effects and future directions. Water Resour. Res. 44 (7).
Computational Tools for Field Implementation of the Florida Phosphorus Senbeta, T.B., Romanowicz, R.J., 2021. The role of climate change and human
Index—Hillsborough County Florida. UF/IFAS Extension. https://citeseerx.ist.psu. interventions in affecting watershed runoff responses. Hydrol. Process. 35 (12),
edu/document?repid=rep1&type=pdf&doi=44a92ab1bcaa747ddb9255961ee7c e14448.
e55da0d9f7a. Shaneyfelt, K.M., Johnson, J.P., Hunt, W.F., 2021. Hydrologic modeling of distributed
ICLUS downloads | US EPA. (2023). . https://www.epa.gov/gcx/about-iclus. stormwater control measure retrofit and examination of impact of subcatchment
Irizarry-Ortiz, M.M., Stamm, J.F., Maran, C., Obeysekera, J., 2022. Development Of discretization in PCSWMM. J. Sustain. Water in the Built Environ. 7 (3), 04021008.
Projected Depth-Duration Frequency Curves (2050–89) for South Florida (2328–0328). Sun, N., Hong, B., Hall, M., 2014. Assessment of the SWMM model uncertainties within
US Geological Survey. the generalized likelihood uncertainty estimation (GLUE) framework for a high-
Jarnagin, S.T., 2007. Historical analysis of the relationship of streamflow flashiness with resolution urban sewershed. Hydrol. Process. 28 (6), 3018–3034.
population density, imperviousness, and percent urban land cover in the Mid- Suttles, K.M., Singh, N.K., Vose, J.M., Martin, K.L., Emanuel, R.E., Coulston, J.W., Saia, S.
Atlantic region. World Environ. Water Res. Congr. 2008, 1–10. Ahupua’A. M., Crump, M.T., 2018. Assessment of hydrologic vulnerability to urbanization and
Jennings, D.B., Taylor Jarnagin, S., 2002. Changes in anthropogenic impervious surfaces, climate change in a rapidly changing watershed in the Southeast US. Sci. Total
precipitation and daily streamflow discharge: a historical perspective in a mid- Environ. 645, 806–816.
Atlantic subwatershed. Landsc. Ecol. 17, 471–489. Tague, C., Pohl-Costello, M., 2008. The Potential Utility of Physically Based Hydrologic
Kadykalo, A., Findlay, C., 2016. The flow regulation services of wetlands. Ecosyst. Serv. Modeling in Ungauged Urban Streams.
20, 91–103. Talbot, M., McGuire, O., Olivier, C., Fleming, R., 2016. Parameterization and application
Khan, S., Deere, D., Leusch, F., Humpage, A., Jenkins, M., Cunliffe, D., 2015. Extreme of agricultural best management practices in a rural Ontario watershed using
weather events: should drinking water quality management systems adapt to PCSWMM. J. Water Manag. Model. C400..
changing risk profiles? Water Res. 85, 124–136. Taylor, K.E., Stouffer, R.J., Meehl, G.A., 2012. An overview of CMIP5 and the experiment
Koepke, J., 2017. Urban stream restoration and applied practices in northeast Illinois. design. Bull. Am. Meteorol. Soc. 93 (4). Article 4.
J. Green Build. 12, 13–27. Urban hydrology for small watersheds. (1986). United States Department of Agriculture.
Lacher, I.L., Ahmadisharaf, E., Fergus, C., Akre, T., Mcshea, W.J., Benham, B.L., Kline, K. USDA, 2019. Web Soil Surv. https://websoilsurvey.nrcs.usda.gov/app/.
S., 2019. Scale-dependent impacts of urban and agricultural land use on nutrients, Villamizar, S.R., Pineda, S.M., Carrillo, G.A., 2019. The effects of land use and climate
sediment, and runoff. Sci. Total Environ. 652, 611–622. change on the water yield of a watershed in Colombia. Water 11 (2), 285.
Marhaento, H., Booij, M.J., Hoekstra, A.Y., 2018. Hydrological response to future land- Wang, R., Kalin, L., Kuang, W., Tian, H., 2014. Individual and combined effects of land
use change and climate change in a tropical catchment. Hydrol. Sci. J. 63 (9), use/cover and climate change on Wolf Bay watershed streamflow in southern
1368–1385. Alabama. Hydrol. Process. 28 (22), 5530–5546.
Martin, G.M., Bellouin, N., Collins, W.J., Culverwell, I., Halloran, P., Hardiman, S.C., Widowati, A., 2017. In: Hydraulic and Hydrologic Modeling of Steep Channel of Putih
Hinton, T., Jones, C.D., McDonald, R.E., McLaren, A.J., O’Connor, F.M., Roberts, M. River, vol. 3, pp. 125–134. Magelang District, Central Java Province, Indonesia.
J., Rodriguez, J.M., Woodward, S., Best, M.J., Brooks, M.E., Brown, A.R., Willems, P., Arnbjerg-Nielsen, K., Olsson, J., Nguyen, V.-T.-V., 2012. Climate change
Butchart, N., Dearden, C., et al., 2011. The HadGEM2 family of met office unified impact assessment on urban rainfall extremes and urban drainage: methods and
model climate configurations, geosci. Model Dev. Discuss. 4, 765–841. shortcomings. Atmos. Res. 103, 106–118.
Meerow, S., Newell, J.P., 2021. Urban resilience for whom, what, when, where, and Wilson, C.O., 2015. Land use/land cover water quality nexus: quantifying anthropogenic
why?. In: Geographic Perspectives on Urban Sustainability. Routledge, pp. 43–63. influences on surface water quality. Environ. Monit. Assess. 187, 1–23.
Moges, E., Demissie, Y., Larsen, L., Yassin, F., 2021. Sources of hydrological model Wirion, C., Bauwens, W., Verbeiren, B., 2017. High Resolution Modeling of the Urban
uncertainties and advances in their analysis. Water 13 (1), 28. Hydrological Response. oint Urban Remote Sensing Event (JURSE), pp. 1–4.
Moniruzzaman, M., Thakur, P.K., Kumar, P., Ashraful Alam, M., Garg, V., Rousta, I., Xian, G., 2007. An analysis of urban development and its environmental impact on the
Olafsson, H., 2020. Decadal urban land use/land cover changes and its impact on Tampa Bay watershed. J. Environ. 85 (4), 965–976.
surface runoff potential for the Dhaka City and surroundings using remote sensing. Xue, D., Zhou, J., Zhao, X., Liu, C., Wei, W., Yang, X., Zhao, Y., 2021. Impacts of climate
Rem. Sens. 13 (1), 83. change and human activities on runoff change in a typical arid watershed, NW
Moreno, M.J., 2011. Analysis of the relationship between submerged aquatic vegetation China. Ecol. Indicat. 121, 107013.
(SAV) and water trophic status of lakes clustered in northwestern Hillsborough Zhou, Q., 2014. A review of sustainable urban drainage systems considering the climate
county, Florida. Water, Air, Soil Pollut. 214, 539–546. change and urbanization impacts. Water 6 (4), 976–992.
Moriasi, D.N., Gitau, M.W., Pai, N., Daggupati, P., 2015. Hydrologic and water quality Zhu, B., Xie, X., Lu, C., Meng, S., Yao, Y., Wang, Y., 2019. Toward high-spatial resolution
models: performance measures and evaluation criteria. Transac. ASABE 58 (6), hydrological modeling for China: calibrating the VIC model. Hydrol. Earth Syst. Sci.
1763–1785. Discuss. 1–43.

14