Monthly estimates of hydrological components were averaged for the early part of the monsoon season from May through July (MJJ), the later part of the monsoon season from August through October (ASO), as well as two other 3-month periods: November through January (NDJ), and February through April (FMA). Trends were determined using the nonparametric
Mann–Kendall trend test, and the corresponding z scores and p values are presented in Table 7. Fig. 7 shows both the average percentage change from long-term average (as percent on left ordinate) and the average quantity (on right ordinate) for the total water yield (mm), soil water content (mm), groundwater recharge (mm), and streamflow (thousand m3 s−1) in four 3-month Omipalisib in vitro periods MJJ, ASO, NDJ and FMA. A significant decreasing trend in the total water yield during MJJ was predicted for the 21st century under both A1B and A2 scenarios with the average water yield remaining
below the baseline ( Fig. 7a). The trend appeared in direct response of the predicted decrease early monsoon precipitation in the basin ( Fig. 6a). Thereafter, increasing trends in the total water yield were predicted for the other periods ( Fig. 7b–d) ( Table 7). The noticeable projection range of total water yield was from 211 mm to 261 mm (5–30% increase from the baseline) during ASO, and it was from 43 mm to 50 mm (20–40% Sirolimus increase from the baseline) during NDJ. In contrast, the long-term patterns Etoposide ic50 of the soil water content showed little change ( Fig. 7e–h) – in the range between 147 mm and 165 mm (3–15% increase from the baseline), which may result from the limited water-holding capacity of the soils ( Wu et al., 2012b). The long-term patterns in the streamflow responded directly to total water yield for the basin. A significant strong decreasing trend in MJJ streamflow was predicted with projection range between 27,525 m3 s−1
and 21,408 m3 s−1 (10–30% decrease from the baseline) ( Fig. 7i) mostly due to predicted decrease in precipitation during the same period. Thereafter, strong increasing trends were detected in the streamflow for the rest of the periods ( Table 7). The projected increase in streamflow ranged from 42,547 m3 s−1 to 55,311 m3 s−1 (0–30% increase from the baseline) during ASO, and 9912–14,372 m3 s−1 (0–45% increase from the baseline) during NDJ under A1B and A2 scenarios, respectively ( Fig. 7j and k). A sharp increasing period in FMA streamflow was also predicted until 2030 primarily possibly due to increased spring snowmelt. The increasing trend followed thereafter, but with much slower rate in the range between 5455 m3 s−1 and 6109 m3 s−1 (0–12% increase from the baseline) ( Fig. 7l). The streamflow patterns during FMA suggested that the impacts of spring snowmelt on the streamflow could diminish by 2030.