Implication of future water regime:
Here we integrated two types of projection: land use and climate data.
Therefore, changes regarding land use and climate are the two main
factors that affected the water regime (Wang, Kalin et al. 2014).
Results show consistency between the moderate and the severe emission
scenarios regarding the projected hydrological variables. Through the
simulation period precipitation increases, consequently, annual
discharge increases. This increase is intensified by growing
urbanization. Land use classes of URMD, URHD, and UIDU change from the
mid-century period to the late-century period by 10%,47%, and 12.5%,
respectively. Therefore, even decreases of precipitation or attenuation
in increase of rainfall towards the end of the century, is compensated
by more increasing impermeable land covers. These land use changes are
along with 23.1% decrease in forest cover (FRSE, FRSD, FRST), 14.7%
hay cover (HAY) and 11.8% agricultural cover (AGRR). RCP4.5 stabilizes
atmospheric radiative forcing at 4.5 W/m2 (650 ppm
CO2 eq) in 2100 (Thomson, Calvin et al. 2011; Van
Vuuren, Edmonds et al. 2011), however, by end of the mid-century period,
the increase in the radiative forcing attenuates significantly. Unlike,
RCP4.5, RCP6.0 keep increasing by end of the century and stabilizes the
radiative forcing at 6.0 W/m2 (850 ppm CO2 eq),
however, GHG emissions declines after the end of the mid-century period
(Van Vuuren, Edmonds et al. 2011). Therefore, under RCP6.0 precipitation
increase during the mid-century and it declines during the late-century
period. As a result, this decline is expected to reflect in discharge
amount, but urban-oriented land use change results in increased
discharge. Similarly, surface runoff and water yield increase in a same
manner. However, since discharge and water yield have linked to
groundwater (Neitsch, Arnold et al. 2011; Arnold, Kiniry et al. 2013),
unlike surface runoff they are indirectly affected. Surface runoff
projection increases are significantly greater for all types of monthly,
seasonal, and annual values. Annual results, for instance, show at least
3 times higher compared to the baseline. Similar studies have concluded
the same result in the region (Wang, Kalin et al. 2014; Sunde, He et al.
2017). Increase in impermeable land covers (URMD, URHD, UIDU) decreases
the amount of infiltration into the soil, and subsequently while
baseflow contribution declines, surface runoff dramatically increases;
this leads to more frequent and intense flooding (Rose and Peters 2001;
Huang, Cheng et al. 2008; Wang, Kalin et al. 2014). Combined effects of
increasing in precipitation, and temperature as well as imperviousness
leads in slight decrease in ET during mid and late century period. Chen
et al. (2017) have reported the same result. This indicates increased
urbanization compensates increased demand of evaporation and
transpiration. Since, the vegetation and tree cover decrease under the
land use scenarios, much less transpiration and plant uptake are
estimated. It can be concluded that the increase in evaporation due to
increase in urbanization cannot be offset by decrease in transpiration.
Also decline in soil water consumption and plant uptake due to less
vegetation-covered lands can lead to increase in stream flow (Price
2011; Sunde, He et al. 2017). Seasonal percent change analysis (figure
10) indicates that most dramatic changes (more frequent extreme
situations) to climate and hydrological variables is projected in the
beginning of mid-century period when switching from the moderate to
severe emissions. Seasonal behavior also agrees with annual changes;
however, it indicates more changes in Winter and Summer. similar results
were reported by Sunde et al. 2017 and wang et al. 2013. In a study
(Georgakakos, Fleming et al. 2014) for the entire southeast region,
decline in storm water has been projected. The general increase in this
study can be attributed to urban-oriented land use change as well as the
watershed specific characteristics like seasonality and storm
frequencies (Villarini and Smith 2010; Sunde, He et al. 2017; Hoyos,
Correa-Metrio et al. 2019). Models also had challenges predicting the
monthly average temperature, monthly average of maximum temperature, and
monthly average of minimum temperature for months of Dec., Jan. and in
some cases February.
Previous studies in the region have used CMIP3 or CMIP5 with general
bias correction. CMIP5 with finer resolution and LOCA with more reliable
climate data, now, have improved the future climate data projections
(Pierce, Cayan et al. 2014; Ficklin, Letsinger et al. 2016). More
realistic regional patterns of precipitation, better estimates of
extreme events, and reduced number of light-precipitation days are the
advantages of LOCA (Pierce, Cayan et al. 2014). These improvements have
reflected more reliable results in this study. This study helped to fill
the a current need to investigate the combined effects of the most
recent downscaled and bias-corrected climate projections and the land
use projections based on SSP (Shared Socioeconomic Pathways) of
Intergovernmental Panel for Climate Change (IPCC). There have been very
few studies of this type investigating the integrated effect of
projected land use and climate data on hydrological responses southeast
US in this study’s scale. UCS is mainly forested and agriculture which
complicates the impacts and responses. More studies are required to
investigate the combined effect of this type of watersheds where notable
level of humidity and proximity to the Golf area which is exposed to
more hurricanes and tropical storms effects the land use and hydrologic
cycle. Wang et al. 2013 have studied an area close to UCS under CMIP3
and concluded the same results of this study. Because of the few number
of research in the region where UCS locates and the also because of the
approach used in this research, the result and projections brought here
can be put in the overall research body and also can be served as a
basis for comparison and decision making process. The approach also can
be utilized for other watersheds to investigate the integrate the land
use and climate projections to study the hydrologic response. A few
Native American Reservations are located within UCS; therefore, this
study can also be used to research the future climate impact on the
reservations’ sustainability and the people. However, it should be
noticed that we used SWAT weather generator to simulate wind speed,
relative humidity, and solar radiation. The soil condition has not been
changed i.e. for all models the current condition was used.