Expected changes in future agro-climatological conditions in Northeast Thailand and their differences between general circulation models

Abstract

We have studied future changes in the atmospheric and hydrological environments in Northeast Thailand from the viewpoint of risk assessment of future cultural environments in crop fields. To obtain robust and reliable estimation for future climate, ten general circulation models under three warming scenarios, B1, A1B, and A2, were used in this study. The obtained change trends show that daily maximum air temperature and precipitation will increase by 2.6°C and 4.0%, respectively, whereas soil moisture will decrease by c.a. 1% point in volumetric water content at the end of this century under the A1B scenario. Seasonal contrasts in precipitation will intensify: precipitation increases in the rainy season and precipitation decreases in the dry season. Soil moisture will slightly decrease almost throughout the year. Despite a homogeneous increase in the air temperature over Northeast Thailand, a future decrease in soil water content will show a geographically inhomogeneous distribution: Soil will experience a relative larger decrease in wetness at a shallow depth on the Khorat plateau than in the surrounding mountainous area, reflecting vegetation cover and soil texture. The predicted increase in air temperature is relatively consistent between general circulation models. In contrast, relatively large intermodel differences in precipitation, especially in long-term trends, produce unwanted bias errors in the estimation of other hydrological elements, such as soil moisture and evaporation, and cause uncertainties in projection of the agro-climatological environment. Offline hydrological simulation with a wide precipitation range is one strategy to compensate for such uncertainties and to obtain reliable risk assessment of future cultural conditions in rainfed paddy fields in Northeast Thailand.

Appendix

1.1 Frequently used abbreviations and symbols

Abbreviation	Description
Pr	Precipitation
ET0	Reference evaporation
Ev	Actual evaporation
Tr	Actual transpiration
Ros	Surface runoff
θ	Volumetric water content (VWC) in soil
θ sat	VWC in soil at saturation (= total porosity of soil)
θ fc	VWC in soil at the field capacity
θ wp	VWC in soil at the wilting point
θ crit	VWC in soil at the critical point ($= \frac{\theta_{\rm fc}+\theta_{\rm wp}}{2}$)

1.2 Evaluation of evapotranspiration

In the FAO-56 framework, evapotranspiration is proportional to the reference evaporation (ET₀) and its efficiency depends on VWC in soil (θ), and the vegetation cover ratio (vg). (In this study, we adopted a very slightly different form of ET₀, with replacement of net radiation with effective input radiation in the original FAO formula.) Evaporation (Ev) and transpiration (Tr) are given by

$${\rm Ev} = K_r(\theta) (1.3 - {\rm Kcb}({\rm vg})) {\rm ET}_0 , $$

(1)

$${\rm Tr} = \sum\limits_{i=2}^{5} {\rm Tr}_i , $$

(2)

$${\rm Tr}_i = {\rm Kcb}({\rm vg}) K_{si}(\theta) {\rm Rp}_i {\rm ET}_0 , $$

(3)

where K _r is the reduction factor under a dry condition and Kcb is the basal crop coefficient function for transpiration under the ideal condition of sufficient water. Generally, Kcb increases with vegetation growth, with a range of 0 ≤ Kcb ≤ 1.3 (because of the complexity of the actual Kcb, this statement is not always true); as a result, evaporation is suppressed whereas transpiration is activated. Transpiration also depends on the water stress factor K _si and the root extension function Rp _i at the ith soil layer. We assumed that plants suck up water from layers 2 to 5.

The factors K _r (θ) and K _s (θ) represent reduction effects of evaporation and transpiration under soil moisture scarcity, respectively. Both factors are modeled by ramp-shaped functions in the FAO-56 paper. In this study, as carried out by Ishigooka et al. (2008), K _s and K _r are evaluated with a fixed parameter (p = 0.5) for simplicity:

$$ K_r = \left\{ \begin{array}{l@{\quad}l} 1 & \theta \ge \theta_{\rm crit} \\ \dfrac{\theta - \frac{1}{2}\theta_{\rm wp}}{\frac{1}{2}\theta_{\rm fc}} & \dfrac{1}{2}\theta_{\rm wp} \le \theta < \theta_{\rm crit} \\ 0 & \theta < \dfrac{1}{2}\theta_{\ \rm wp} \end{array} \right. $$

(4)

and

$$ K_s = \left\{ \begin{array}{l@{\quad}l} 1 & \theta \ge \theta_{\rm crit} \\ \dfrac{\theta - \theta_{\rm wp}}{\theta_{\rm crit} - \theta_{\rm wp}} & \theta_{\rm wp} \le \theta < \theta_{\rm crit} \\ 0 & \theta < \theta_{\rm wp} , \end{array} \right. $$

(5)

where θ _fc, θ _wp, and $\theta_{\rm crit} (= \frac{\theta_{\rm fc} + \theta_{\rm wp}}{2})$ are the VWC in soil at the field capacity, wilting point and “critical point,” respectively.