Figure 8. Number of hot days in default (DEF) and wetland scheme (WS) simulations and the reduction in hot days from WS to DEF.
Figure 9 shows the wetland cooling effect on temperature versus its maximum saturated fraction for the domain. The \(F_{\text{satmx}}\)parameter generally corresponds to highly saturated regions with high water availability. The cooling effect from wetlands is evident for almost every month in the simulation period and is stronger in June and July than in May and August. The strongest effect is shown in 2005 July of almost 1.5 cooler. There exists a linear relationship between\(F_{\text{satmx}}\) and TEMP, especially in 2005 and 2007 — the larger the \(F_{\text{satmx}}\), the stronger the cooling effect is. This pattern is not as clear in 2006, indicating that the wetland cooling effect relies on the available water input from precipitation: under normal and wet conditions, the greater the \(F_{\text{satmx}}\), the more water can be stored in wetland storage, the more surface energy partitioned into latent heat, hence the stronger the cooling effect. However, in the 2006 dry conditions, there is not sufficient precipitation to fill these wetlands, limiting the wetlands’ cooling effects and evaporation, regardless of \(F_{\text{satmx}}\) values. This wetland cooling effect is analogous to the classic soil moisture-temperature feedback in land-atmosphere interactions (Senviratnes et al., 2010; Perkins, 2015).