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.