1. Introduction
Climate models have become an important tool for climate simulation (Pathak et al. 2021), extreme weather and climate events diagnosis (Huang et al. 2021), and climate change prediction (Ghose et al. 2021). With the help of numerical models, researchers have enhanced understanding of climate change and revealed many climate change mechanisms that are difficult to obtain by traditional means (Ham et al. 2021). Numerical modeling is constantly expanding human understanding of Earth’s climate system. The results of this research have been applied to formulating responses to the projected future changes of Earth’s climate system (Tebaldi et al. 2021). However, numerical model simulation results also have a certain degree of uncertainty, the magnitude of which is affected by unavoidable defects of the numerical model itself and observation errors. The difference between physical parameterization schemes is an essential source of climate simulation uncertainty (Hou et al. 2012).
Soil thermal conductivity (STC) is one of the key parameters of the coupled numerical model of water, heat, and solutes in soil physics. As the key physical model for calculating LST in land surface models (LSM), STC controls the heat conduction process in soils under steady state (Hou et al. 2012; Peters-Lidard et al. 1998). And it can influence physical, biological and chemical processes through regulating energy partitioning at the ground surface and energy distribution at subsurface soil layers (Peters-Lidard et al. 1998; Ochsner et al. 2001; Pan et al. 2017). With phase change of soil water occurring, the STC also affects soil freeze–thaw cycles and soil water movement (Cuntz and Haverd 2018; Wang and Yang 2018). Global comparisons show the similar relationships that STC significantly affects the simulated soil temperature and other related thermal and hydraulic variables over arid and semiarid regions in mid and high latitudes (Dai et al. 2019). Researches have proposed many indirect estimation models to describe the relationship between STC and soil texture, bulk density, and water content (Devries 1963; Tong et al. 2009; Zhang et al. 2017). There are two kinds of indirect estimation models: theoretical models and empirical models. Johansen (1977) proposed the concept of normalized thermal conductivity and studied the effects of soil type, porosity (n), saturation (Sr), and mineral composition on STC, establishing the Johansen semiempirical and semitheoretical STC model, which is widely used in the current LSMs. Lu et al. (2007) overcomes the defect of Johansen’s model that does not consider the soil water content enough, and established a new model based on the linear relationship between the thermal conductivity of dry soil and the porosity of mineral soil. The model partly solves the limitation of soil moisture content on the calculation of STC, and can simulate the thermal conductivity in a wide range of soil moisture content. The simulation accuracy of soil temperature is higher, and the simulation bias is smaller than that of other models.
Located in the mid- to high latitudes of Asia, northern China is one of the major arid and semiarid regions in the world, accounting for more than half of China’s land area. The Gobi Desert is distributed in this region and the land use also includes grassland, farmland, glacier, and a small amount of forest. With the gradual enhancement of the East Asia summer monsoon (EASM) and South Asian monsoon activities, the water vapor transported inland from the ocean increases from April, reaches maximum from June to August, and gradually weakens in September (Wang et al. 2007). As a result of these changes in water vapor transport, April to September is the main period of precipitation in northern China, accounting for more than 80% of the annual precipitation (Zhang and Qian 2003). Therefore, the amount of precipitation in this period determines not only the dry and wet conditions in northern China but also the change trend of drought, making it the most important factor affecting aridification of the region. However, the surface vegetation and soil properties in northern China are easily affected by human activities (Deng et al. 2018). As a result of the strong land–atmosphere interaction in this region, the climate effect caused by the change of surface properties is strong (Zhang et al. 2020). The STC is a key factor reflecting the change of underlying surface; however, the effects of STC on the uncertainty of rainy season precipitation simulation in this region remain unclear. Therefore, the study of this issue is important for the improvement of LSM performance, enhancement of the simulation ability of climate change, and revealing the mechanisms of climate change.
In this paper, the regional climate model RegCM4.6 coupled with the latest generation land surface model CLM4.5 is used to reveal the impact and mechanism of STC change on rainy season precipitation in northern China, to further understanding of the regional climate change mechanism. In addition, it provides a new basis for making decisions to deal with the impact of human activities on future climate change.
2. Study area and data
a. Study area
The study area of this paper includes northeast, north, and northwest China (Fig. 1). Affected by the EASM, the eastern part of this region has relatively high precipitation levels. The surface vegetation is mainly forest and farmland with some meadow shrubs and high vegetation coverage. Forest is mainly distributed in northeast Inner Mongolia, northeast China, and southern Shaanxi. The center and west of the study area includes central Inner Mongolia, northern Gansu, and Xingjiang, which are usually controlled by the westerly belt and are located far from the ocean. Meanwhile, the center and western part the study area is blocked by the Qinghai–Tibetan Plateau. Water vapor from southern China does not easily penetrate this region. Consequently, the atmosphere is dry with rare precipitation and the underlying surface is mainly desert, the Gobi Desert, and sparse vegetation.
The topographic height (m) in the simulation region and locations of provinces of northern China. The point marks are precipitation observing stations.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0003.1
The topographic height (m) in the simulation region and locations of provinces of northern China. The point marks are precipitation observing stations.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0003.1
The topographic height (m) in the simulation region and locations of provinces of northern China. The point marks are precipitation observing stations.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0003.1
b. Data sources
In this study, the daily observed precipitation and LST data from 377 meteorological stations in northern China (Fig. 1) were selected, and the data from 1 January 1988 to 31 December 2017 were used.
3. Research methods
a. Soil thermal conductivity
b. Surface energy balance equation
c. Model and numerical test scheme
RegCM is a regional climate model established in the late 1980s by Dickinson et al. (1989) and Giorgi et al. (1993) after expanding and modifying the radiation scheme, convection parameterization scheme, and land surface physical process in the mesoscale model MM4. As currently the most widely used regional climate model in China, RegCM is not only used for climate simulation and diagnosis research but also as a climate prediction business support tool (Gao et al. 2017). RegCM4.6 is the latest mature version, adding the MM5 nonstatic dynamic framework option. Moreover, the radiation parameterization scheme and convection parameterization scheme were updated in this version. RegCM4.6 was coupled with the land surface model CLM4.5. CLM4.5 was developed on the basis of the second-generation LSMs, including BATS, IAP94, and NCAR-LSM. The model has been through several iterations including CLM2.0, CLM3.0, CLM3.5, and CLM4.0 (Dickinson et al. 2006). Compared with previous versions, CLM4.5 contained a revised photosynthesis scheme, improved hydrological process and wetland distribution in cold regions, a new snow cover parameterization scheme, lake model, crop model, and various urban categories, and a new biological nitrogen fixation mechanism and methane emission model in the vertical direction for soil.
This paper uses the regional climate model RegCM4.6 coupled with the land surface model CLM4.5. The following simulation parameters were used in this study. The simulation area was 15.76°–57.36°N, 66.25°–141.13°E, which covers northern China (Fig. 1). There were 130 grid points in the latitudinal direction and 345 grid points in the meridional direction. In addition, the horizontal grid distance was 30 km and there were 23 uneven layers in the vertical direction. The mixed coordinate system was adopted. The top air pressure was 50 hPa. The Arakawa–Lamb B-type staggered grid was adopted in the horizontal direction. The projection was Lambert projection. The MM5 nonhydrostatic equilibrium dynamic framework was used in the experiments. The other main physical parameterization schemes are shown in Table 1. All parameters have been shown to be effective by many previous studies (Oh et al. 2014; Gao et al. 2017; Ren et al. 2018). The latest global atmospheric reanalysis data, the ERA-Interim, were used for the model side boundary field data. It is 0.75° × 0.75° horizontal resolution and has 37 vertical levels. These data come from the European Centre for Medium-Range Weather Forecasts (ECMWF). And the quality of this dataset is globally recognized and is often used to initialize numerical weather forecasts and climate models (Balsamo et al. 2015).
Primary physical process schemes.
Using the above parameters, two numerical experiments named CLM-JH and CLM-LR were conducted. In CLM-JH, the Johansen scheme of STC was adopted, whereas in CLM-LR, the Lu–Ren scheme of STC was adopted. The two experiments were, respectively, integrated for 31 years (1 January 1987–31 December 2017), of which 1987 was the spinup period (Dai et al. 2019; Wang et al. 2016; Gao et al. 2016) and the following 30 years (January 1988–December 2017) were the analysis period.
d. Pearson correlation coefficient
4. Sensitivity analysis of the simulation results
The observational analysis in the study period showed that the average precipitation from April to September in northern China was lower in the northwest and higher in the southeast and increased from northwest to southeast. The precipitation level in most regions of Xingjiang, Qinghai, and western Gansu was less than 150 mm, while that in southern Xingjiang basin, the region with the least precipitation in northern China, was less than 50 mm. The precipitation in regions affected by the edge of EASM such as north China and northeast China exceeded 500 mm (Fig. 2a). Compared with the observation, CLM-JH and CLM-LR were able to simulate the spatial distribution and variation characteristics of precipitation (Figs. 2b,c). The precipitation bias of CLM-JH was large, at 199.8 mm. However, the precipitation bias simulated by CLM-LR was significantly reduced, at only 91.2 mm. The precipitation difference of the two tests can reflect the degree of sensitivity of the precipitation simulation. Figure 2d illustrates that the region with the greatest sensitivity was within the EASM zone on the eastern edge of the study area, including Heilongjiang, Jilin, and eastern Liaoning, where the precipitation difference exceeded 200 mm. The western study area includes arid regions such as northern Xingjiang, Gansu, western Inner Mongolia, and northern Qinghai and the precipitation difference exceeded 100 mm in this area.
Observed and simulation of annual average precipitation (mm) from April to September in northern China. (a) Observed, (b) CLM-JH, (c) CLM-LR, and (d) precipitation of CLM-LR minus that of CLM-JH.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0003.1
Observed and simulation of annual average precipitation (mm) from April to September in northern China. (a) Observed, (b) CLM-JH, (c) CLM-LR, and (d) precipitation of CLM-LR minus that of CLM-JH.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0003.1
Observed and simulation of annual average precipitation (mm) from April to September in northern China. (a) Observed, (b) CLM-JH, (c) CLM-LR, and (d) precipitation of CLM-LR minus that of CLM-JH.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0003.1
The analysis of monthly observed precipitation showed that July had highest precipitation levels, reaching 110 mm, followed by August and June with 90 and 70 mm, respectively. April had the lowest precipitation levels, reaching only 30 mm (Fig. 3a). Both experiments were able to simulate the variation characteristics of monthly precipitation. The deviation between simulated and observed precipitation was smallest in July and August but largest in April. Compared with CLM-JH, the simulated precipitation bias of CLM-LR was smaller and closer to the observed precipitation. The monthly precipitation difference simulated by the two schemes gradually increased from April, reaching the maximum in June, and gradually decreased in July, August, and September (Fig. 3a). Correspondingly, the EASM began to affect the region in April, had the strongest effect in June and July, and weakened in September, indicating that the intensity of the monsoon effect is a key factor affecting the sensitivity of precipitation simulation in the region.
Average precipitation from April to September in northern China. (a) Monthly precipitation and (b) annual precipitation (mm).
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0003.1
Average precipitation from April to September in northern China. (a) Monthly precipitation and (b) annual precipitation (mm).
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0003.1
Average precipitation from April to September in northern China. (a) Monthly precipitation and (b) annual precipitation (mm).
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0003.1
The analysis of observed precipitation showed that the rainy season precipitation in northern China increased over the study period at a rate of 2.9 mm (10a)−1. This trend did not reach the statistically significant level of P = 0.05 for its trend coefficient was 0.08 (Fig. 3b). CLM-JH simulated the interannual precipitation change well; the correlation coefficient was 0.61, reaching t-test significance level of a = 0.001 (the t-test critical correlation coefficient of a = 0.001 is around 0.55). However, CLM-JH did not simulate the rising trend of precipitation in the study period [−5.7 mm (10a)−1]. CLM-LR simulated the precipitation variation trend [2.9 mm (10a)−1], with a correlation coefficient of 0.73, which was a stronger correlation compared with that of CLM-JH.
5. Analysis of the causes of differences between two precipitation simulations
a. Relationship between LST and precipitation variation
The LST simulated by CLM-LR was generally higher than that of CLM-JH. The Xingjiang, Gansu, and Inner Mongolia regions generally showed a LST difference of more than 1 K. Northern Tianshan, southwest Xingjiang, the northern Qilian Mountains, and central and western Inner Mongolia showed a LST difference of more than 2 K (Fig. 4a). There was a negative correlation between precipitation and LST variations. Moreover, the region of significant correlation (Fig. 4b) corresponded to the region with a significant increase of LST. (Fig. 4a). When the local LST increases by 1 K, precipitation decreases by 5–30 mm in most areas of northern China and more than 30 mm in Qinghai Plateau and Northeast China except for narrow areas of 112°–120°E. (Fig. 4c). There was an opposite trend between annual precipitation change and LST change. The correlation coefficient was −0.42, reaching the t-test significance level of a = 0.005 (Fig. 4d). These findings suggest that an increase of LST leads to a decrease of precipitation. When the LST increases by 1 K, the precipitation decreases most in the eastern and southern edges of the study area. The reason is that these regions are located in the transition zone of the EASM, and the increase of LST not only affects the change of local precipitation conditions, but also causes the contraction of the EASM and the reduction of water vapor transport (see related analysis content below). Other regions in the study area are far away from the EASM, and the increase of LST only affects the local precipitation conditions. Therefore, the decrease of precipitation is basically linear with the increase of LST.
Relationship between LST and precipitation in the rainy season (April–September). (a) LST change and (b) correlation between precipitation and LST change, where the color bar levels correspond to the significance levels, that is, 0.35 corresponds to the significance level at 0.05, 0.41 corresponds to 0.01, and 0.55 corresponds to 0.001, and so on. (c) Ratio of CLM-LR and CLM-JH simulated precipitation difference to surface temperature difference (mm K−1), and (d) annual changes of regional average precipitation and LST.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0003.1
Relationship between LST and precipitation in the rainy season (April–September). (a) LST change and (b) correlation between precipitation and LST change, where the color bar levels correspond to the significance levels, that is, 0.35 corresponds to the significance level at 0.05, 0.41 corresponds to 0.01, and 0.55 corresponds to 0.001, and so on. (c) Ratio of CLM-LR and CLM-JH simulated precipitation difference to surface temperature difference (mm K−1), and (d) annual changes of regional average precipitation and LST.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0003.1
Relationship between LST and precipitation in the rainy season (April–September). (a) LST change and (b) correlation between precipitation and LST change, where the color bar levels correspond to the significance levels, that is, 0.35 corresponds to the significance level at 0.05, 0.41 corresponds to 0.01, and 0.55 corresponds to 0.001, and so on. (c) Ratio of CLM-LR and CLM-JH simulated precipitation difference to surface temperature difference (mm K−1), and (d) annual changes of regional average precipitation and LST.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0003.1
b. Adjustment of surface energy balance factors
The analysis in section 5a showed an anticorrelated relationship between LST change and precipitation. The LST change affects the surface net radiation (refers to the net downward shortwave energy flux) through the change of longwave radiation energy, resulting in a change in the distribution of sensible heat (refers to the surface upward sensible heat flux) and latent heat. The surface net radiation is below 400 W m−2 in the west of the north China, 400–450 W m−2 in the north of and above 450 in the southeast of the region (Fig. 5a). The greatest reduction of surface net radiation occurred in southwest Xingjiang, northern Tianshan in Xingjiang, the Hexi Corridor in Gansu, and central to western Inner Mongolia with a reduction value of 8–12 W m−2, corresponding to the regions with the greatest increase in LST (Fig. 5b). The surface sensible heat flux is more than 40 W m−2 in the western desert area and 10–30 W m−2 in other areas (Fig. 5c). The increase in LST increased the sensible heat flux. The sensible heat flux increased by about 2 W m−2 on average from April to September. The regions with the largest increase corresponded to those with the greatest increase in LST (around 8 W m−2), including southwest Xingjiang, northern Tianshan in Xingjiang, Hexi Corridor in Gansu, and central and western Inner Mongolia. Southwest Xingjiang and western Hexi in Gansu had the greatest increases of around 12 W m−2 (Fig. 5d). The latent heat flux is more than 400 (W m−2, the same below) and less than 400 from the east and from the west of Inner Mongolia boundary, respectively. In addition, it is less than 350 in the plateau area (Fig. 5e). The decrease of surface net radiation (approximately equal to the sum of sensible and latent heat fluxes) and the increase of sensible heat flux led to a decrease of surface latent heat flux of more than 2 W m−2 in northern China. The regions with the greatest reduction were southwest Xingjiang, northern Tianshan in Xingjiang, Hexi Corridor in Gansu, and central and western Inner Mongolia. The reduction value for these regions was above 10 W m−2 and up to 18 W m−2 in the regions of highest reduction, including the junction of Xingjiang, Gansu, Inner Mongolia, and southwest Xingjiang (Fig. 5f).
The surface incident shortwave radiation, sensible heat flux, latent heat flux simulated by CLM-JH, and their differences between CLM-LR and CLM-JH in the rainy season (April–September). (a) Surface incident shortwave radiation simulated by CLM-JH, (b) surface incident shortwave radiation of CLM-LR minus that of CLM-JH, (c) surface sensible heat flux simulated by CLM-JH, (d) surface sensible heat flux of CLM-LR minus that of CLM-JH, (e) surface latent heat flux simulated by CLM-JH, and (f) surface sensible heat flux of CLM-LR minus that of CLM-JH.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0003.1
The surface incident shortwave radiation, sensible heat flux, latent heat flux simulated by CLM-JH, and their differences between CLM-LR and CLM-JH in the rainy season (April–September). (a) Surface incident shortwave radiation simulated by CLM-JH, (b) surface incident shortwave radiation of CLM-LR minus that of CLM-JH, (c) surface sensible heat flux simulated by CLM-JH, (d) surface sensible heat flux of CLM-LR minus that of CLM-JH, (e) surface latent heat flux simulated by CLM-JH, and (f) surface sensible heat flux of CLM-LR minus that of CLM-JH.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0003.1
The surface incident shortwave radiation, sensible heat flux, latent heat flux simulated by CLM-JH, and their differences between CLM-LR and CLM-JH in the rainy season (April–September). (a) Surface incident shortwave radiation simulated by CLM-JH, (b) surface incident shortwave radiation of CLM-LR minus that of CLM-JH, (c) surface sensible heat flux simulated by CLM-JH, (d) surface sensible heat flux of CLM-LR minus that of CLM-JH, (e) surface latent heat flux simulated by CLM-JH, and (f) surface sensible heat flux of CLM-LR minus that of CLM-JH.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0003.1
c. Variation of local precipitation conditions
Sensible heat, latent heat, and the change of surface evapotranspiration affect the stability and humidity of the lower atmosphere, thus affecting the local precipitation conditions. The wet static energy in CLM-JH decreases from 3.3 (×105 J kg−1) in the southwest to less than 3.1 in the northeast of north China (Fig. 6a). Figure 7b shows the difference of the moist static energy between the CLM-LR and the CLM-JH. In CLM-LR, the moist static energy in northern China declined by about 0.2 × 103 K hPa−1 compared to that of CLM-JH. The regions of greatest reduction were again located in southwest Xingjiang, northern Tianshan in Xingjiang, Hexi Corridor in Gansu, and central and western Inner Mongolia, with the highest increase in LST exceeding 0.6 × 103 K hPa−1 (Fig. 6b). In CLM-JH, the uplift condensation level increases from 850 hPa in northeast of the region to 600 hPa in southeast of the region (Fig. 6c). Figure 6d shows the difference of the uplift condensation level between CLM-LR and the CLM-JH. With the decrease of the moist static energy, the lifting condensation height increased by about 10 hPa. In southwest Xingjiang, northern Tianshan in Xingjiang, Hexi Corridor in Gansu, and central and western Inner Mongolia, the lifting condensation height increased by more than 30 hPa (Fig. 6d). These findings indicate that the increase of LST forms local thermal and water vapor conditions that are not conducive to precipitation.
The moist static energy and lifting condensation level simulated by CLM-JH and its differences between CLM-LR and CLM-JH in rainy season (April–September). (a) Moist static energy simulated by CLM-JH, (b) moist static energy difference between CLM-LR and CLM-JH, (c) lifting condensation level simulated by CLM-JH, and (d) lifting condensation level difference between CLM-LR and CLM-JH.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0003.1
The moist static energy and lifting condensation level simulated by CLM-JH and its differences between CLM-LR and CLM-JH in rainy season (April–September). (a) Moist static energy simulated by CLM-JH, (b) moist static energy difference between CLM-LR and CLM-JH, (c) lifting condensation level simulated by CLM-JH, and (d) lifting condensation level difference between CLM-LR and CLM-JH.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0003.1
The moist static energy and lifting condensation level simulated by CLM-JH and its differences between CLM-LR and CLM-JH in rainy season (April–September). (a) Moist static energy simulated by CLM-JH, (b) moist static energy difference between CLM-LR and CLM-JH, (c) lifting condensation level simulated by CLM-JH, and (d) lifting condensation level difference between CLM-LR and CLM-JH.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0003.1
Variation of 500-hPa subtropical high in the rainy season (April–September). (a) 500-hPa geopotential height simulated by CLM-LR and CLM-JH, (b) geopotential height difference between CLM-LR and CLM-JH, (c) along the 115°E–time profile, and (d) along the 25°N–time profile (10 gpm).
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0003.1
Variation of 500-hPa subtropical high in the rainy season (April–September). (a) 500-hPa geopotential height simulated by CLM-LR and CLM-JH, (b) geopotential height difference between CLM-LR and CLM-JH, (c) along the 115°E–time profile, and (d) along the 25°N–time profile (10 gpm).
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0003.1
Variation of 500-hPa subtropical high in the rainy season (April–September). (a) 500-hPa geopotential height simulated by CLM-LR and CLM-JH, (b) geopotential height difference between CLM-LR and CLM-JH, (c) along the 115°E–time profile, and (d) along the 25°N–time profile (10 gpm).
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0003.1
d. Variation in the 500-hPa circulation system
The 500-hPa subtropical high is one of the most important circulation factors affecting precipitation in northern China from April to September (Wu and Liu 1992). Liu et al. (2001) showed that the change of surface evapotranspiration has an important effect on the occurrence and development of the subtropical high. The 500-hPa isobaric surface situation from April to September in mainland China showed that Xingjiang is below a weak ridge; the southeast is dominated by the subtropical high, while the northeast is dominated by a low pressure trough (Fig. 7a). Figure 7a also indicated that the edge of the west pacific subtropical high (the contour 588 dagpm) of CLM-LR is farther south than that of CLM-JH. Figure 7b shows the geopotential height difference field of the 500-hPa isobaric surface before and after the increase of LST, and indicates that the geopotential height in northern China increases after the increase of LST from April to September. This also indicates that the Xingjiang ridge is strengthened and the northeast low pressure trough is weakened, resulting in the strengthening of the divergent downdraft across northern China. The geopotential height field in southeast China was negative, indicating that the subtropical high system is weakened. This climatic system makes it more difficult for water vapor to be transported to inland northern China, thus forming a large-scale circulation field which is not conducive to precipitation (Fig. 7b).
The LST change also has a major effect on the activity of the subtropical high ridge. Figures 7c and 7d show geopotential height time profiles along the blue line given in Figs. 7a. Figure 7c shows that after the LST increased, the northward movement of the subtropical high 588-dagpm equipotential line was greatly weakened, retreating around 1° southward from June to August. Furthermore, the eastward retreat of the subtropical high 588-dagpm equipotential line is also clearly apparent. The time taken to reach the same longitude from April to September was also delayed relative to normal (Fig. 7d). These findings suggest that the LST can cause changes to the large-scale circulation system by affecting the change of surface thermal factors, forming a circulation field that is unfavorable to precipitation.
e. The influence of STC change on wind field at 500 hPa
According to the analysis of 500-hPa wind simulated by CLM-JH, the whole Chinese mainland is controlled by the westerlies from April to September, with wind speed above 8 m s−1 in most regions of northern China, and above 10 m s−1 in east of 100°E (Fig. 8a). Compared with that simulated by CLM-JH, the results simulated by CLM-LR showed a weaker west wind in eastern part of 100°E, a narrower wind speed band that above 10 m s−1 and enhanced Westerly winds in the western part of 100°E (Fig. 8b). Wind field differences between the CLM-JH and CLM-LR shows that the there is an anomalous east wind zone from the east of 100°E, a west wind zone from the west of 100°E, and an enhanced anticyclone circulation in Qinghai Plateau (Fig. 8c).
The 500-hPa wind simulated by CLM-JH and CLM-LR: (a) CLM-JH, (b) CLM-LR, and (c) wind of CLM-LR minus that of CLM-JH.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0003.1
The 500-hPa wind simulated by CLM-JH and CLM-LR: (a) CLM-JH, (b) CLM-LR, and (c) wind of CLM-LR minus that of CLM-JH.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0003.1
The 500-hPa wind simulated by CLM-JH and CLM-LR: (a) CLM-JH, (b) CLM-LR, and (c) wind of CLM-LR minus that of CLM-JH.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0003.1
f. The effect of STC change on water vapor transport at 700 hPa
Simulated vapor flux at 700 hPa by CLM-JH shows that a vapor transport belt is located at 25°N from April to September, and the vapor flux value is between 0.03 and 0.04 g s−1 hPa−1 cm−1. In some areas of southeast China, the value is above 0.04 g s−1 hPa−1 cm−1 (Fig. 9a). However, in CLM-LR, the belt swings slightly to the southeast and its value decreases to 0.02–0.03 g s−1 hPa−1 cm−1 (Fig. 9b). Compared the vapor flux value of two experiments, In CLM-LR, except for a few regions, the value decreases in most regions of northern China, especially in the areas affected by the East Asian summer monsoon by more than 0.004 g s−1 hPa−1 cm−1 (Fig. 9c).
The 700-hPa water vapor flux simulated by CLM-JH and CLM-LR: (a) CLM-JH, (b) CLM-LR, and (c) difference between CLM-LR and CLM-JH.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0003.1
The 700-hPa water vapor flux simulated by CLM-JH and CLM-LR: (a) CLM-JH, (b) CLM-LR, and (c) difference between CLM-LR and CLM-JH.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0003.1
The 700-hPa water vapor flux simulated by CLM-JH and CLM-LR: (a) CLM-JH, (b) CLM-LR, and (c) difference between CLM-LR and CLM-JH.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0003.1
g. The influence of STC on local precipitation conditions
According to the vapor flux divergence of CLM-JH at 700 hPa (Fig. 10a), except for part of the west and middle of north China with weak divergence, most of the area has convergence from April to September. And the most significant convergence area is east and northeast of Inner Mongolia, in which the divergence value reaches −0.01 g s−1 hPa−1 cm−1. In CLM-LR, the vapor subsidence area expanded and its convergence value decreased (Fig. 10b). It can be concluded that the change of STC increases LST, resulting in the increase of the subsidence degree of the 700-hPa vapor flux, which is not conducive to the generation of precipitation (Fig. 10c).
The 700-hPa water vapor flux divergence simulated by CLM-JH and CLM-LR (g s−1 hPa−1 cm−1): (a) CLM-JH, (b) CLM-LR, and (c) difference between CLM-LR and CLM-JH.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0003.1
The 700-hPa water vapor flux divergence simulated by CLM-JH and CLM-LR (g s−1 hPa−1 cm−1): (a) CLM-JH, (b) CLM-LR, and (c) difference between CLM-LR and CLM-JH.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0003.1
The 700-hPa water vapor flux divergence simulated by CLM-JH and CLM-LR (g s−1 hPa−1 cm−1): (a) CLM-JH, (b) CLM-LR, and (c) difference between CLM-LR and CLM-JH.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0003.1
h. The interactions of the land surface feedbacks with the East Asian summer monsoon system
To study the influence of the EASM on land surface, the soil water content (SWC) in the whole soil layer was analyzed. As shown in Fig. 11a, the SWC in CLM-JH is below 900 kg m−2 in the desert region of northern China, while in most of the East Asian monsoon system affected areas east of 104°E it exceeds 1100 kg m−2. The SWC of the whole layer simulated by CLM-LR and its difference with CLM-JH shows the southeast retreat of the East Asian summer monsoon, which reduces the SWC of the whole of North China (Fig. 11b). Among them, the SWC in the area affected by the EASM generally decreased by more than 50 kg m−2 (Fig. 11c).
Soil water content of the whole soil layer simulated by CLM-JH and CLM-LR (kg m−2): (a) CLM-JH, (b) CLM-LR, and (c) soil water content of CLM-LR minus that of CLM-JH.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0003.1
Soil water content of the whole soil layer simulated by CLM-JH and CLM-LR (kg m−2): (a) CLM-JH, (b) CLM-LR, and (c) soil water content of CLM-LR minus that of CLM-JH.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0003.1
Soil water content of the whole soil layer simulated by CLM-JH and CLM-LR (kg m−2): (a) CLM-JH, (b) CLM-LR, and (c) soil water content of CLM-LR minus that of CLM-JH.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0003.1
i. Influencing mechanisms
Based on the above analysis, we propose the mechanism of the effect of LST change on precipitation presented in Fig. 12. First, the change of surface properties leads to the anomaly of LST, which further results in change in the surface energy balance and the distribution of sensible and latent heat. The changes in latent heat then change the surface evapotranspiration, affect the change of lower atmospheric stability and water content, and form local thermal and water vapor conditions that are not conducive to precipitation. Finally, the surface evapotranspiration change affects the change of the 500-hPa circulation system, resulting in local and large-scale conditions that are unfavorable to precipitation. The black dotted line shows that the abnormal reduction of precipitation can lead to the degradation of surface vegetation and land surface aridification in northern China, change the physical properties of surface soil such as STC, and then affect the variations of precipitation.
Mechanism behind the effect of land surface temperature change on precipitation. TCPW: total column precipitable water, MSE: moisture static energy, LCL: lifting condensation level, SR: subtropical ridge, LR: longwave radiation, SH: surface sensible heat flux, SNR: surface net radiation, and LH: latent heat flux.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0003.1
Mechanism behind the effect of land surface temperature change on precipitation. TCPW: total column precipitable water, MSE: moisture static energy, LCL: lifting condensation level, SR: subtropical ridge, LR: longwave radiation, SH: surface sensible heat flux, SNR: surface net radiation, and LH: latent heat flux.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0003.1
Mechanism behind the effect of land surface temperature change on precipitation. TCPW: total column precipitable water, MSE: moisture static energy, LCL: lifting condensation level, SR: subtropical ridge, LR: longwave radiation, SH: surface sensible heat flux, SNR: surface net radiation, and LH: latent heat flux.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0003.1
6. Discussion
We refer to some studies about simulate summer precipitation using RegCM4 in areas affected by the East Asian monsoon. These studies adopted different LSMs of BATS and CLM and different convection parameterization schemes (Chen et al. 2014; Bao 2013; Wang et al. 2014; Li and Ding 2004; Wang et al. 2016), and even adopted WRF mesoscale models. The results show that these experiments simulate precipitation more generally than the observed precipitation in this area. After analysis, we found that the low simulated temperature and high evaporation of these models are an important reason for excessive precipitation in this region, and their influence mechanism has been elaborated in detail. According to the analysis of formulas (2)–(4), the difference between the two LSTs is mainly Ke function form. In the Johansen scheme, Ke is only a function of soil moisture. The type of underlying surface in northern China changes rapidly from west to east. The Johansen model does not fully conform to the fact of the underlying surface in northern China, which may be an important reason for the error. In the Lu–Ren scheme, the influence of soil texture is considered, so the simulation bias is reduced compared with the Johansen scheme, especially in extremely arid areas. In the two sensitivity experiments of STC carried out in this paper, the improvement of the LST cold deviation from the Lu–Ren scheme in the arid and semiarid areas in northern China was better than that of East Asian summer monsoon, so the deviation improvement of precipitation simulation is more obvious. It shows that the calculation model adopted in the Lu–Ren scheme is more suitable for the regions with low soil moisture, while most regions in northern China belong to arid and semiarid regions, with scarce precipitation and low soil moisture, so the simulation effect of the Lu–Ren scheme is good.
On the other hand, since the industrial revolution, the rapid increase of global population and the social and economic activities have led to land surface aridification including decrease of global forests and increase of farmland and urban areas. The land surface aridification resulting in the abnormal exchange of substances and energy between the land and atmosphere. As a result, the local climate was significantly changed and the extreme weather and climatic events increased (IPCC 2021).
Table 2 shows the LST variation trend of different periods of northern China. The results showed that the increase trend of overall LST is very significant in the recent 30 years in northern China. The most significant temperature increase period is winter, with a range of 1.36 K (10a)−1, reaching the significance level of 0.001. So, land surface aridification is a very important variation in the northern part of China. It may cause changes in STC. A change in STC leads to a change in LST. And finally it affects regional precipitation through land–atmosphere interaction.
Trends of LST in different periods [K (10a)−1]. Here, three asterisks (***) represents a significance level above 0.001.
The anomaly of surface net radiation energy caused by changes in surface albedo is another important cause of local climate change. In the 1970s, Otterman first proposed the influence of land–atmosphere interaction on drought. He attributed the long-term drought in the Sahel region to the bare surface caused by human grazing (Otterman 1974). Charney first proposed a dynamic interpretation of the impact of land–atmosphere interaction on drought by Otterman: the Sahel is in the sinking area of the Hadley circulation, the climate is dry and rainless, and the surface damage increases the albedo and reflects more solar radiation, forming a radiation heat sink compared with the surrounding area. To maintain the thermal balance, the air sinks in this area, thereby strengthening the sinking area of the Hadley circulation and aggravating the drought. This further degraded the vegetation (Charney 1975).
No matter the change of surface albedo or STC, it will change the latent heat of surface by changing the distribution of net radiation energy or surface energy, and then cause the change of the local geothermal and dynamic structure of the atmosphere. Changes in local circulation are superimposed on the large-scale circulation, which results in variations in precipitation.
7. Conclusions
This study examined the impact of STC on the simulation of rainy season precipitation in northern China and revealed the mechanism of the impact of LST change on precipitation using the regional climate model RegCM4.6 coupled with the latest generation land surface model CLM4.5, through the long-term numerical experiments of two STC models. It was found that the STC has a significant effect on the simulated precipitation from April to September in northern China, especially in the influence area of the EASM in the east of the study area. Moreover, the increase of LST caused by the change of soil thermal conductivity reduce precipitation. When the local LST increases by 1 K, precipitation decreases by 5–30 mm in most areas of northern China. The abnormal LST changes the distribution of surface energy and changes the stability and water content of the lower atmosphere, thus affecting the 500-hPa circulation system and the 700-hPa vapor transportation and its divergence, forming local and large-scale circulation conditions that are not conducive to precipitation.
The results of this paper not only reveal the mechanism of the change of underlying surface properties affecting the rainy season precipitation in northern China, but also add a new soil thermal conductivity scheme for the new-generation land surface model CLM4.5, which improves the simulation effect of northern China. The results also provide a decision-making basis for formulating the climate effects of human activities in the region.
Acknowledgments.
This research was funded by the National Natural Science Foundation of China (42230611, U2142208, 41975016); Natural Science Foundation of Gansu Province (Grants 20JR5RA119, 20JR5RA112); Opening Research Foundation of Key Laboratory of Land Surface Process and Climate Change in Cold and Arid Regions, Chinese Academy of Sciences (LPCC2020001); National Key Research and Development Program of China (2020YFA060840X); Opening Foundation of Key Laboratory of Desert and Desertification, Chinese Academy of Sciences (KLDD-2020-01); the National Natural Science Foundation of China (Grant 41805079); and Innovation Team of Gansu Meteorological Bureau (Grant GSQXCXTD-2020-01).
Data availability statement.
The authors declare no conflict of interest. Climatic data were downloaded from http://data.cma.cn/user/toLogin.html.
REFERENCES
Balsamo, G., and Coauthors, 2015: ERA-Interim/Land: A global land surface reanalysis dataset. Hydrol. Earth Syst. Sci., 19, 389–407, https://doi.org/10.5194/hess-19-389-2015.
Bao, Y., 2013: Simulations of summer monsoon climate over East Asia with a Regional Climate Model (RegCM) using Tiedtke convective parameterization scheme (CPS). Atmos. Res., 134, 35–44, https://doi.org/10.1016/j.atmosres.2013.06.009.
Barnes, S. L., 1994: Applications of the Barnes objective analysis scheme. Part I: Effects of undersampling, wave position, and station randomness. J. Atmos. Oceanic Technol., 11, 1433–1448, https://doi.org/10.1175/1520-0426(1994)011<1433:AOTBOA>2.0.CO;2.
Charney, J. G., 1975: Dynamics of deserts and drought in the Sahel. Quart. J. Roy. Meteor. Soc., 101, 193–202, https://doi.org/10.1002/qj.49710142802.
Chen, J., Y. Yin, Q. Chen, H. Ding, and H. Xiao, 2014: Effect of convection schemes on the simulation of monsoon climates: A sensitivity study using RegCM4. Climate Res., 60, 147–162, https://doi.org/10.3354/cr01229.
Cuntz, M., and V. Haverd, 2018: Physically accurate soil freeze-thaw processes in a global land surface scheme. J. Adv. Model. Earth. Syst., 10, 54–77, https://doi.org/10.1002/2017MS001100.
Dai, Y., N. Wei, H. Yuan, S. Zhang, W. Shangguan, S. Liu, X. Lu, and Y. Xin, 2019: Evaluation of soil thermal conductivity schemes for use in land surface modeling. J. Adv. Model. Earth Syst., 11, 3454–3473, https://doi.org/10.1029/2019MS001723.
Deng, C. H., and Coauthors, 2018: Spatial- temporal variation of the vegetation coverage in Qinling Mountains and its dual response to climate change and human activities. J. Nat. Resour., 33, 425–438.
DeVries, D. A., 1963: Thermal properties of soils. Physics of Plant Environment, W. R. van Wijk, Ed., North-Holland Publishing Company, 210–235.
Dickinson, R. E., R. M. Errico, F. Giorgi, and G. T. Bates, 1989: A regional climate model for the western United States. Climatic Change, 15, 383–422, https://doi.org/10.1007/BF00240465.
Dickinson, R. E., K. W. Oleson, G. Bonan, F. Hoffman, P. Thornton, M. Vertenstein, Z.-L. Yang, and X. Zeng, 2006: The community land model and its climate statistics as a component of the Community Climate System Model. J. Climate, 19, 2302–2324, https://doi.org/10.1175/JCLI3742.1.
Gao, X., Y. Shi, and F. Giorgi, 2016: Comparison of convective parameterizations in RegCM4 experiments over China with CLM as the land surface model. Atmos. Oceanic Sci. Lett., 9, 246–254, https://doi.org/10.1080/16742834.2016.1172938.
Gao, X., Y. Shi, Z. Han, M. Wang, J. Wu, D. Zhang, Y. Xu, and F. Giorgi, 2017: Performance of RegCM4 over major river basins in China. Adv. Atmos. Sci., 34, 441–455, https://doi.org/10.1007/s00376-016-6179-7.
Ghose, B., A. R. M. T. Islam, M. Kamruzzaman, M. Moniruzzaman, and Z. Hu, 2021: Climate-induced rice yield anomalies linked to large-scale atmospheric circulation in Bangladesh using multi-statistical modeling. Theor. Appl. Climatol., 144, 1077–1099, https://doi.org/10.1007/s00704-021-03584-2.
Giorgi, F., M. R. Marinucci, G. T. Bates, and G. DeCanio, 1993: Development of a second-generation regional climate model (RegCM2): Convective processes and assimilation of lateral boundary conditions. Mon. Wea. Rev., 121, 2814–2832, https://doi.org/10.1175/1520-0493(1993)121<2814:DOASGR>2.0.CO;2.
Grell, G. A., 1993: Prognostic evaluation of assumptions used by cumulus parameterizations. Mon. Wea. Rev., 121, 764–787, https://doi.org/10.1175/1520-0493(1993)121<0764:PEOAUB>2.0.CO;2.
Ham, Y.-G., Y.-S. Joo, and J.-Y. Park, 2021: Mechanism of skillful seasonal surface chlorophyll prediction over the Southern Pacific using a global earth system model. Climate Dyn., 56, 45–64, https://doi.org/10.1007/s00382-020-05403-2.
Holtslag, A. A. M., E. I. F. de Bruijn, and H-L. Pan, 1990: A high resolution air mass transformation model for short-range weather forecasting. Mon. Wea. Rev., 118, 1561–1575, https://doi.org/10.1175/1520-0493(1990)118<1561:AHRAMT>2.0.CO;2.
Hou, Z., M. Huang, L. R. Leung, G. Lin, and D. M. Ricciuto, 2012: Sensitivity of surface flux simulations to hydrologic parameters based on an uncertainty quantification framework applied to the community land model. J. Geophys. Res., 117, D15108, https://doi.org/10.1029/2012JD017521.
Huang, W.-R., Y.-H. Chang, L. Deng, and P.-Y. Liu, 2021: Simulation and projection of summer convective afternoon rainfall activities over Southeast Asia in CMIP6 models. J. Climate, 34, 5001–5016, https://doi.org/10.1175/JCLI-D-20-0788.1.
IPCC, 2021: Future global climate: Scenario-42 based projections and near-term information. Climate Change 2021: The Physical Science Basis, V. Masson-Delmotte et al., Eds., Cambridge University Press, 553–672.
Johansen, O., 1977: Thermal conductivity of soils. Tech. Rep. CRREL TL 637, Trondheim University Norway, 291 pp.
Kiehl, J. T., J. J. Hack, G. B. Bonan, B. A. Boville, B. P. Briegleb, D. L. Williamson, and P. J. Rasch, 1996: Description of the NCAR Community Climate Model (CCM3). NCAR Tech. Note NCAR/TN-420+STR, 152 pp., https://doi.org/10.5065/D6FF3Q99.
Li, Q.-P., and Y.-H. Ding, 2004: Multi-year simulation of the East Asian monsoon and precipitation in China using a regional climate model and evaluation. Acta Meteor. Sin., 62, 140–153, https://doi.org/10.11676/qxxb2004.016.
Liu, Y. M., G. X. Wu, H. Liu, and P. Liu, 2001: Condensation heating of the Asian summer monsoon and the subtropical anticyclone in the Eastern Hemisphere. Climate Dyn., 17, 327–338, https://doi.org/10.1007/s003820000117.
Lu, S., T. Ren, Y. Gong, and R. Horton, 2007: An improved model for predicting soil thermal conductivity from water content at room temperature. Soil Sci. Soc. Amer. J., 71, 8–14, https://doi.org/10.2136/sssaj2006.0041.
Ochsner, T. E., R. Horton, and T. Ren, 2001: A new perspective on soil thermal properties. Soil Sci. Soc. Amer. J., 65, 1641–1647, https://doi.org/10.2136/sssaj2001.1641.
Oh, S.-G., J.-H. Park, S.-H. Lee, and M.-S. Suh, 2014: Assessment of the RegCM4 over East Asia and future precipitation change adapted to the RCP scenarios. J. Geophys. Res., 119, 2913–2927, https://doi.org/10.1002/2013JD020693.
Otterman, J., 1974: Baring high-albedo soils by overgrazing: A hypothesized desertification mechanism. Science, 186, 531–533, https://doi.org/10.1126/science.186.4163.531.
Pal, J. S., E. E. Small, and E. A. B. Eltahir, 2000: Simulation of regional-scale water and energy budgets: Representation of subgrid cloud and precipitation processes within RegCM. J. Geophys. Res., 105, 29 579–29 594, https://doi.org/10.1029/2000JD900415.
Pan, Y., S. Lyu, S. Li, Y. Gao, X. Meng, Y. Ao, and S. Wang, 2017: Simulating the role of gravel in freeze–thaw process on the Qinghai–Tibet Plateau. Theor. Appl. Climatol., 127, 1011–1022, https://doi.org/10.1007/s00704-015-1684-7.
Pathak, R., S. Sahany, and S. K. Mishra, 2021: Impact of stochastic entrainment in the NCAR CAM deep convection parameterization on the simulation of South Asian summer monsoon. Climate Dyn., 57, 3365–3384, https://doi.org/10.1007/s00382-021-05870-1.
Peters-Lidard, C. D., E. Blackburn, X. Liang, and E. F. Wood, 1998: The effect of soil thermal conductivity parameterization on surface energy fluxes and temperatures. J. Atmos. Sci., 55, 1209–1224, https://doi.org/10.1175/1520-0469(1998)055<1209:TEOSTC>2.0.CO;2.
Ren, Y., Y. Li, Z. Pu, T. Zhang, H. Duan, and W. Wang, 2018: Effects of updated RegCM4 land use data on near-surface temperature simulation in China. J. Meteor. Res., 32, 758–767, https://doi.org/10.1007/s13351-018-7156-0.
Tebaldi, C., and Coauthors, 2021: Climate model projections from the scenario model intercomparison project (ScenarioMIP) of CMIP6. Earth Syst. Dyn., 12, 253–293, https://doi.org/10.5194/esd-12-253-2021.
Tong, F. G., L. R. Jing, and R. W. Zimmerman, 2009: An effective thermal conductivity model of geological porous medium for coupled thermo-hydro-mechanical systems with multiphase flow. Int. J. Rock Mech. Min. Sci., 46, 1358–1369, https://doi.org/10.1016/j.ijrmms.2009.04.010.
Wang, C., and K. Yang, 2018: A new scheme for considering soil water-heat transport coupling based on Community Land Model: Model description and preliminary validation. J. Adv. Model. Earth Syst., 10, 927–950, https://doi.org/10.1002/2017MS001148.
Wang, D.-H., and Coauthors, 2007: Advances in the study of rainstorm in Northeast China. Diqiu Kexue Jinzhan, 22, 549–560, https://doi.org/10.11867/j.issn.1001-8166.2007.06.0549.
Wang, X., M. Yang, and G. Pang, 2014: Sensitivity of regional climate simulations to land-surface schemes on the Tibetan Plateau. Climate Res., 62, 25–43, https://doi.org/10.3354/cr01262.
Wang, X., G. Pang, M. Yang, and G. Wan, 2016: Effects of modified soil water-heat physics on RegCM4 simulations of climate over the Tibetan Plateau. J. Geophys. Res. Atmos., 121, 6692–6712, https://doi.org/10.1002/2015JD024407.
Wu, G., and H. Liu, 1992: Atmospheric precipitation in response to equatorial and tropical sea surface tempreature anomalies. J. Atmos. Sci., 49, 2236–2255, https://doi.org/10.1175/1520-0469(1992)049<2236:APIRTE>2.0.CO;2.
Zeng, X., M. Zhao, and R. E. Dickinson, 1998: Intercomparison of bulk aerodynamic algoriths for the computation of sea surface fluxes using TOGA COARE and TAO data. J. Climate, 11, 2628–2644, https://doi.org/10.1175/1520-0442(1998)011<2628:IOBAAF>2.0.CO;2.
Zhang, L.-J., and Y.-F. Qian, 2003: Annual distribution features of precipitation in China and their interannual variations. Acta Meteor. Sin., 17, 146–163.
Zhang, N., X. Yu, A. Pradhan, and A. J. Puppala, 2017: A new generalized soil thermal conductivity model for sand-kaolin clay mixtures using thermo-time domain reflectometry probe test. Acta Geotech., 12, 739–752, https://doi.org/10.1007/s11440-016-0506-0.
Zhang, Q., J. Zeng, P. Yue, L. Zhang, S. Wang, and R. Wang, 2020: On the land-atmosphere interaction in the summer monsoon transition zone in East Asia. Theor. Appl. Climatol., 141, 1165–1180, https://doi.org/10.1007/s00704-020-03239-8.
