Verification of Planetary Boundary Layer Height for Local Data Assimilation and Prediction System (LDAPS) Using the Winter Season Intensive Observation Data during ICE-POP 2018

ICE-POP 2018기간 동계집중관측자료를 활용한 국지수치모델(LDAPS)의 행성경계층고도 검증

Abstract

Planetary boundary layer height (PBLH), produced by the Local Data Assimilation and Prediction System (LDAPS), was verified using RawinSonde (RS) data obtained from observation at Daegwallyeong (DGW) and Sokcho (SCW) during the International Collaborative Experiments for Pyeongchang 2018 Olympic and Paralympic winter games (ICE-POP 2018). The PBLH was calculated using RS data by applying the bulk Richardson number and the parcel method. This calculated PBLH was then compared to the values produced by LDAPS. The PBLH simulations for DGW and SCW were generally underestimation. However, the PBLH was an overestimation from surface to 200 m and 450 m at DGW and SCW, respectively; this result of model's failure to correctly simulate the Surface Boundary Layer (SBL) and the Mixing Layer (ML) as the PBLH. When the accuracy of the PBLH simulation is low, large errors are seen in the mid- and low-level humidity. The highest frequencies of Planetary boundary layer (PBL) types, calculated by the LDAPS at DGW and SCW, were presented as types Ι and II, respectively. Analysis of meteorological factors according to the PBL types indicate that the PBLH of the existing stratocumulus were overestimated when the mid- and low-level humidity errors were large. If the instabilities of the surface and vertical mixing into clouds are considered important factors affecting the estimation of PBLH into model, then mid- and low-level humidity should also be considered important factors influencing PBLH simulation performance.

Fig. 1. Spatial location and information on the rawinsonde observation sites and LDAPS grid (black soild lines).

Fig. 2. Scatter plot of planetary boundary layer height from rawinsonde vs. LDAPS in a) Daegwallyeong and b) Sokcho. c) cumulative distribution function of planetary boundary layer height from rawinsonde (OBS) and LDAPS (LDPS) in Daegwallyeong and Sokcho.

Fig. 3. Vertical profiles of virtual potential temperature (θ_v), mixing ratio (Q), saturation mixing ratio (Qs), Wind speed (WS), and Richardson number on clear day (a, c) and before precipitation (b, d) from rawinsonde in Daegwallyeong (a, b) and Sokcho (c, d). In this figures, yellow and blue cross dot line is PBLH from LDAPS and rawinsonde, respectively.

Fig. 4. Diurnal mean values of PBLH from rawinsonde and LDAPS. The bar is bias mean between LDAPS and rawinsonde.

Fig. 5. Vertical profile of average (solid line; rawinsonde, dashed line; LDAPS) of temperature, relativity humidity, and wind speed on the a) absolute error less than 10%, b) absolute error greater than 90% in Daegwallyeong and Sokcho (c; ~10%, d; 90%~).

Fig. 6. The frequency of planetary boundary layer type from LDAPS.

Fig. 7. Vertical profile of average(solid line; rawinsonde, dashed line; LDAPS) of temperature, relativity humidity, and wind speed by planetary boundary layer type a) 1, b) 2, c) 3, d) 4, e) 5, f) 6, and g) 7 in Daegwallyeong.

Fig. 8. Same as Fig. 7 in Sokcho.

Table 1. Specifications of rawinsonde (M10).

Table 2. Diurnal mean, bias mean and RMSE of PBLH estimated from rawinsonde and LDAPS in Daegwallyeong and Sokcho.

Table 3. Bias and RMSE of temperature, relative humidity, and wind speed separated by PBLH absolute error range between rawinsonde and LDAPS in Daegwallyeong.

Table 4. Same as Table 2 in Sokcho.

Table 5. Summary of LDAPS boundary layer types 1 to 7 (from Boutle et al., 2015).

Table 6. Average, bias and RMSE of PBLH by PBL type in Daegwallyeong and Sokcho.

Table 7. Bias and RMSE of temperature, relative humidity and wind speed by PBL type in Daegwallyeong.

Table 8. Same as Table 4 in Sokcho.