Due to the important “barrier” effect of the Himalayas on moisture transport and precipitation over the TP, previous studies have examined the impacts of topographical complexity of the Himalayas by analyzing the simulation results with different horizontal resolutions. Although different horizontal resolutions can resolve different scales of topographical complexity, they can also introduce differences in physical processes such as convective clouds. Therefore, this study examines the modeling difference between the V4km and V4km-smooth experiments to reflect the impacts of topographical complexity as discussed below. The convection-permitting resolution at the refinement region covering the entire TP can guarantee reasonably simulated meteorological fields during the SASM season as discussed above and resolve the topographical complexity to a large extent.
Figure 10 shows the spatial distributions of integrated moisture transport of the V4km over the TP region averaged for JJA of 2015 and the difference between V4km and V4km.smooth. The shaded contour represents the magnitude of moisture transport. It is obvious that the Himalayas diverts the WVT to the TP into two branches, one passage through the southwestern TP and the other through the Yarlung Tsangpo Grand Canyon into the southeastern TP. To quantify the moisture transport through the different pathways into the TP, this study defines the TP as the region within the black box in Fig. 2. The five dashed lines denote the TP’s five boundaries (B1–B5). The estimates of column integrated moisture transport through the five boundaries into the TP averaged for June-August of 2015 are listed in Table 1. It is evident that moisture is transported away from the TP (negative values) through B1 and B2, while it is transported into the TP (positive values) through B3, B4, and B5. The net effect is for moisture to be transported into the TP. Moisture is mainly transported into the TP through B3, contributing about ~61% to the total moisture transported into the TP in JJA. The moisture transported from B4 and B5 is comparable, contributing ~23% and ~16%, respectively. The pathway through B2 (~99.8%) dominates the moisture transport away from the TP.
Figure 10. Spatial distributions of integrated water vapor transport for the simulation with the complex topography averaged from 1 June to 31 August 2015, and the difference between the simulations with the complex and smooth topography.
Moisture transport (Tg h−1) V4km V4km.smooth B1 –0.2 –10.9 B2 –103.3 –124.6 B3 98.9 117.9 B4 38.0 44.2 B5 26.1 27.4 Sum 59.5 54.0
Table 1. Integrated moisture transport through the five boundaries (denoted in Fig. 2) into the TP. 1 Tg = 1012 g
The impact of topographical complexity, i.e., the difference between V4km and V4km.smooth, weakens the moisture transport through the higher mountains and enhances the moisture transport through the deeper valleys such as the Yarlung Tsangpo Grand Canyon (Figs. 10 and S2 in ESM). The overall moisture transport toward the TP through B3, B4, and B5 is weakened by the topographical complexity (Fig. 10), which reduces the moisture transport from 117.9 Tg h–1 to 98.9 Tg h–1, from 44.2 Tg h–1 to 38.0 Tg h–1, and from 27.4 Tg h–1 to 26.1 Tg h–1 through B3, B4, and B5, respectively (Table 1). This reduction of moisture transport through the Himalayas is mainly due to the complex topography increasing the surface roughness and weakening the wind fields, consistent with previous studies (e.g., Lin et al., 2018; Wang et al., 2020). The analysis shows that the impact of topographical complexity on moisture transport is mainly below 500 hPa (Fig. S2 in ESM). With a global variable-resolution simulation, in addition to the moisture transport across the Himalayas (B3–B5), the transport through other boundaries (B1–B2) can also be examined. Figure 10 shows that the moisture transport away the TP through B1 and B2 is also weakened by the complex topography. The moisture transport away from the TP is reduced from 10.9 Tg h–1 to 0.2 Tg h–1 and from 124.6 Tg h–1 to 103.3 Tg h–1 through B1 and B2, respectively. Therefore, the net effect of complex topography on moisture transport into the TP is positive overall, increasing the transport from 54.0 Tg h–1 to 59.5 Tg h–1. This result indicates that although the complex topography weakens the moisture transport through the Himalayas, its overall effect increases the net moisture transport into the TP.
Besides the moisture transport, Table 2 shows the moisture budget terms over the TP as denoted in Fig. 2. The impact on the moisture transport term has been discussed above, i.e., the topographical complexity increases the net moisture flow into the TP by ~11% (from 54.0 Tg h–1 to 59.5 Tg h–1). The evaporation rate over the TP displays a small increase of ~2% (from 103.6 Tg h–1 to 105.8 Tg h–1). The precipitation over the TP is enhanced by ~3% from 186.2 Tg h–1 to 192.2 Tg h–1. Therefore, in general, the impacts of topographical complexity on the moisture budget terms over the entire TP are primarily reflected by the moisture transport term, while the impacts on other terms are relatively small. Although the analysis seems to indicate that the impact of topographical complexity on the precipitation of entire TP is small, some previous modeling studies demonstrated that the complex topography might modulate the precipitation around the Himalayas through various other mechanisms, for example, orographic drag (e.g., Karki et al., 2017; Wang et al., 2020). Therefore, the impacts of complex topography on the precipitation are analyzed, focusing on the Himalayas region below.
Moisture budget (Tg h−1) V4km V4km.smooth WVT 59.5 54.0 Evaporation 105.8 103.6 Precipitation 192.2 186.2
Table 2. Moisture budget over the TP. 1 Tg = 1012 g.
Figure 11 shows the spatial distributions of difference in precipitation between the V4km and V4km.smooth simulations averaged for JJA of 2015. It shows that the primary impacts of topographical complexity on precipitation concentrate on the Himalayas region, while the precipitation changes within the TP are small. The Himalayan region is further divided into three subregions, western (blue box), central (black box), and eastern (red box) (Fig. 11) for the detailed analysis. The average precipitation over the three subregions is shown in Table 3. Over the western Himalayas, the average precipitation from the V4km simulation is 9.56 mm d–1, slightly less than that from V4km.smooth (9.78 mm d–1). Over the central Himalayas, precipitation is reduced by about 11% from 11.82 mm d–1 (in V4km) to 10.56 mm d–1 (in V4km.smooth). Over the eastern Himalayas, the difference between V4km (14.67 mm d–1) and V4km.smooth (14.69 mm d–1) is quite small. Although the difference of average precipitation over the western and eastern Himalayas is relatively small, it is evident that the spatial distributions are significantly modulated by the topographical complexity, which is further investigated below.
Figure 11. Spatial distributions of the difference in precipitation between the simulations with the complex and smooth topography averaged from 1 June to 31 August 2015, with the blue rectangular region being the western Himalayas, the black rectangular region being the central Himalayas and the red rectangular region being the eastern Himalayas. Table 3 shows the average precipitation in these areas.
Average precipitation (mm d−1) V4km V4km.smooth Western 9.56 9.78 Central 11.82 10.56 Eastern 14.67 14.69
Table 3. The average precipitation over the three Himalayan sub-regions: western, central, and eastern (denoted by blue, black, and red boxes, respectively, in Fig. 11). 1 Tg = 1012 g.
Figure 12 shows the precipitation amounts and the terrain heights along the direction perpendicular to the Himalayas for V4km and V4km.smooth averaged over the three subregions. The difference in terrain height between V4km and V4km.smooth is also shown. Over all subregions, precipitation from both simulations mainly concentrates along the slope, with the maxima located at the lower levels of the slope. Precipitation decreases sharply in the upslope direction, consistent with the spatial pattern shown in Fig. 9. Generally, over the western Himalayas, higher terrain leads to higher precipitation in the V4km compared to V4km.smooth. The reduction of precipitation in V4km (e.g., 30.3°N to 30.7°N and 31.2°N to 31.6°N) compared to V4km.smooth corresponds well with its higher terrain nearby, while the increase occurs where the elevation is higher in the V4km (e.g., 30.7°N to 31.2°N).
Figure 12. The precipitation amounts and elevation along the direction perpendicular to the Himalayas for two simulations with complex and smooth topography averaged over the three subregions (denoted by blue, black, and red boxes, respectively, in Fig. 11), from 1 June to 31 August 2015. The X-axis of “western” is the latitude along the northwestern boundary of the blue box in Fig. 11. Blue solid (dashed) lines denote precipitation in V4km (V4km.smooth). Black denotes the elevation difference between two simulations in top plots and denotes elevation from V4km in bottom plots. Red dashed lines in three subregions denote the location of transects analyzed in Fig. 13.
Furthermore, precipitation changes are generally located south of the terrain difference between the two experiments. Over the central Himalayas, the relationship between the differences in precipitation and terrain from the two experiments is similar. The impacts of topographical complexity on precipitation over these two regions are mainly due to the narrower and sharper slopes (in the direction perpendicular to the Himalayas) resolved in V4km, which have the effect of shifting the lifted airflow, and hence precipitation, northward compared to V4km.smooth. Rahimi et al. (2019) also found this northward precipitation shift upon comparing two experiments at different horizontal resolutions. However, over the eastern Himalayas, the abovementioned mechanism seems less evident compared to the other central and western regions. This is mainly due to the flow through the Yarlung Tsangpo Grand Canyon exerting a significant impact on moisture transport and precipitation over this region, noting that the canyon is better resolved in V4km than in V4km.smooth. Therefore, the precipitation differences between the two experiments, i.e., higher precipitation in V4km, are mainly attributed to this resolved canyon.
Aside from the precipitation along the direction perpendicular to the Himalayas, the difference in precipitation along the Himalayas over the three regions (denoted by three red dashed lines in Fig. 12) between the two experiments is also analyzed (Fig. 13). Over the western Himalayas, there is a strong correlation between precipitation and terrain height in V4km. The peak precipitation corresponds with mountains and less precipitation occurs in the small-scale valleys (e.g., ~79.8°E, ~81.7°E, and ~82.5°E). However, in V4km.smooth, precipitation is generally higher than in V4km because its terrain is smoothed, and thus the small-scale valleys are not resolved well. Precipitation in V4km.smooth is also higher without fully resolving the valleys over the central (e.g., 92.5°E to 93.2°E) and the eastern (e.g., 95.9°E to 97.1°E, and 97.3°E to 98°E) Himalayas. Therefore, it is evident that the greater number of valleys parallel to the moisture transport across the Himalayas, better resolved by V4km, serve as channels for moisture transport and favor the northward shift of precipitation compared to V4km.smooth.
Figure 13. The precipitation amounts and elevation along the Himalayas (denoted by three red dashed lines in Fig. 12) of two simulations with complex and smooth topography averaged from 1 June to 31 August 2015. Blue solid (dashed) lines denote precipitation in V4km (V4km.smooth), and the black lines denote elevation.
|Moisture transport (Tg h−1)|