Author Contributions
Conceptualization, C.Y. and H.W.; methodology, C.Y.; software, C.Y.; validation, C.X.; formal analysis, C.Y.; investigation, H.W.; resources, C.Y.; data curation, C.Y.; writing—original draft preparation, C.Y.; writing—review and editing, C.X.; visualization, C.Y.; supervision, H.W.; project administration, C.X.; funding acquisition, C.Y. All authors have read and agreed to the published version of the manuscript.
Figure 1.
Topography and distribution of meteorological stations in Shandong Province, China. A total of 109 meteorological stations in the region from the China Meteorological Data Center were collected with a temporal resolution of 1 h. The hourly ERA5 atmospheric temperature data at 2 m above the land surface in Shandong Province were downloaded and cropped, with a total of 328 grid points. The Advanced Spaceborne Thermal Emission and Reflection Radiometer Global Digital Elevation Model (ASTER GDEM) was used to minimize the height difference between the reanalysis data and the actual meteorological observations.
Figure 1.
Topography and distribution of meteorological stations in Shandong Province, China. A total of 109 meteorological stations in the region from the China Meteorological Data Center were collected with a temporal resolution of 1 h. The hourly ERA5 atmospheric temperature data at 2 m above the land surface in Shandong Province were downloaded and cropped, with a total of 328 grid points. The Advanced Spaceborne Thermal Emission and Reflection Radiometer Global Digital Elevation Model (ASTER GDEM) was used to minimize the height difference between the reanalysis data and the actual meteorological observations.
Figure 2.
Temperature residual sequences for the four different meteorological stations under the three schemes. The latitude (°), longitude (°) and elevation (m) of (a) (36.25, 117.10, 1533.70), (b) (37.32, 120.85, 205.50), (c) (35.43, 119.53, 37.10) and (d) (36.42, 115.97, 34.30).
Figure 2.
Temperature residual sequences for the four different meteorological stations under the three schemes. The latitude (°), longitude (°) and elevation (m) of (a) (36.25, 117.10, 1533.70), (b) (37.32, 120.85, 205.50), (c) (35.43, 119.53, 37.10) and (d) (36.42, 115.97, 34.30).
Figure 3.
Difference sequences of adjacent epochs for the atmospheric temperature between ERA5 and meteorological stations from 1 October 1 to 31 December 2022. The latitude (°), longitude (°) and elevation (m) of (a) (36.25, 117.10, 1533.70), (b) (37.32, 120.85, 205.50), (c) (35.43, 119.53, 37.10) and (d) (36.42, 115.97, 34.30).
Figure 3.
Difference sequences of adjacent epochs for the atmospheric temperature between ERA5 and meteorological stations from 1 October 1 to 31 December 2022. The latitude (°), longitude (°) and elevation (m) of (a) (36.25, 117.10, 1533.70), (b) (37.32, 120.85, 205.50), (c) (35.43, 119.53, 37.10) and (d) (36.42, 115.97, 34.30).
Figure 4.
Best-fit polynomials. (a) UTC 1 h (top); (b) UTC 9 h.
Figure 4.
Best-fit polynomials. (a) UTC 1 h (top); (b) UTC 9 h.
Figure 5.
Bias sequences of temperature at UTC 1 h before and after model correction at four different elevation meteorological stations. The latitude (°), longitude (°) and elevation (m) of (a) (36.25, 117.10, 1533.70), (b) (37.32, 120.85, 205.50), (c) (35.43, 119.53, 37.10) and (d) (36.42, 115.97, 34.30).
Figure 5.
Bias sequences of temperature at UTC 1 h before and after model correction at four different elevation meteorological stations. The latitude (°), longitude (°) and elevation (m) of (a) (36.25, 117.10, 1533.70), (b) (37.32, 120.85, 205.50), (c) (35.43, 119.53, 37.10) and (d) (36.42, 115.97, 34.30).
Figure 6.
Bias sequences of temperature at UTC 9 h before and after model correction at four different elevation meteorological stations. The latitude (°), longitude (°) and elevation (m) of (a) (36.25, 117.10, 1533.70), (b) (37.32, 120.85, 205.50), (c) (35.43, 119.53, 37.10) and (d) (36.42, 115.97, 34.30).
Figure 6.
Bias sequences of temperature at UTC 9 h before and after model correction at four different elevation meteorological stations. The latitude (°), longitude (°) and elevation (m) of (a) (36.25, 117.10, 1533.70), (b) (37.32, 120.85, 205.50), (c) (35.43, 119.53, 37.10) and (d) (36.42, 115.97, 34.30).
Figure 7.
Residual sequence of the meteorological stations at different elevations before and after the RRM refinement. The latitude (°), longitude (°) and elevation (m) of (a) (36.25, 117.10, 1533.70), (b) (37.32, 120.85, 205.50), (c) (35.43, 119.53, 37.10) and (d) (36.42, 115.97, 34.30).
Figure 7.
Residual sequence of the meteorological stations at different elevations before and after the RRM refinement. The latitude (°), longitude (°) and elevation (m) of (a) (36.25, 117.10, 1533.70), (b) (37.32, 120.85, 205.50), (c) (35.43, 119.53, 37.10) and (d) (36.42, 115.97, 34.30).
Figure 8.
Bias sequences at the interpolated meteorological stations for ERA5 2 m atmospheric temperature under the four schemes with increasing meteorological station elevations.
Figure 8.
Bias sequences at the interpolated meteorological stations for ERA5 2 m atmospheric temperature under the four schemes with increasing meteorological station elevations.
Figure 9.
Residual sequence of the PWV at the meteorological stations. (a) 54826; (b) 54836; (c) 54945.
Figure 9.
Residual sequence of the PWV at the meteorological stations. (a) 54826; (b) 54836; (c) 54945.
Table 1.
The average RMS of the temperature residual for all meteorological stations with different elevations under the three schemes.
Table 1.
The average RMS of the temperature residual for all meteorological stations with different elevations under the three schemes.
Elevation Interval (m) | Refinement Strategy |
---|
Solution A (K) | Solution B (K) | Solution C (K) |
---|
1533 | 5.28 | 3.56 | 1.98 |
150~310 | 2.07 | 1.66 | 1.57 |
100~150 | 1.72 | 1.45 | 1.43 |
50~100 | 1.33 | 1.21 | 1.18 |
0~50 | 1.09 | 1.01 | 1.01 |
Table 2.
Average RMS of temperature bias at UTC 1 h and UTC 9 h before (Uncorr.) and after (Corr.) model correction for different elevation intervals (K).
Table 2.
Average RMS of temperature bias at UTC 1 h and UTC 9 h before (Uncorr.) and after (Corr.) model correction for different elevation intervals (K).
Elevation Interval (m) | UTC 1 h | UTC 9 h |
---|
Uncorr. | Corr. | Uncorr. | Corr. |
---|
1533 | 5.40 | 2.35 | 3.68 | 2.06 |
150~310 | 4.07 | 2.18 | 3.37 | 1.87 |
100~150 | 3.32 | 1.99 | 3.10 | 1.65 |
50~100 | 2.88 | 1.66 | 2.66 | 1.53 |
0~50 | 2.81 | 1.59 | 2.45 | 1.36 |
Table 3.
Average STD of the residual series of grid temperature at different elevation intervals before and after the RRM refinement.
Table 3.
Average STD of the residual series of grid temperature at different elevation intervals before and after the RRM refinement.
Elevation Interval (m) | Corrections without Using the RRM (K) | Corrections Using the RRM (K) | Enhancement Rate (%) |
---|
1533 | 1.06 | 0.76 | 28.3 |
150~310 | 0.84 | 0.69 | 17.9 |
100~150 | 0.76 | 0.65 | 14.5 |
50~100 | 0.68 | 0.59 | 13.2 |
0~50 | 0.65 | 0.58 | 10.8 |
Table 4.
Average RMS at the interpolated meteorological stations for atmospheric temperature under different elevation intervals.
Table 4.
Average RMS at the interpolated meteorological stations for atmospheric temperature under different elevation intervals.
Elevation (m) | Solution A (K) | Solution B (K) | Solution C (K) | Solution D (K) | Solution E (K) |
---|
1533 | 5.28 | 1.98 | 1.06 | 0.76 | 5.79 |
150~310 | 2.04 | 1.55 | 0.83 | 0.58 | 2.37 |
100~150 | 1.69 | 1.44 | 0.76 | 0.54 | 1.87 |
50~100 | 1.31 | 1.16 | 0.73 | 0.53 | 1.39 |
0~50 | 1.08 | 1.02 | 0.69 | 0.48 | 1.16 |
Table 5.
Average RMS at interpolated meteorological stations for the under the four schemes for different elevation intervals.
Table 5.
Average RMS at interpolated meteorological stations for the under the four schemes for different elevation intervals.
Elevation Interval (m) | Solution A (K) | Solution B (K) | Solution C (K) | Solution D (K) |
---|
1533 | 4.28 | 1.60 | 0.82 | 0.62 |
150~310 | 1.68 | 1.27 | 0.68 | 0.48 |
100~150 | 1.39 | 1.16 | 0.63 | 0.45 |
50~100 | 1.08 | 0.96 | 0.60 | 0.42 |
0~50 | 0.88 | 0.82 | 0.58 | 0.41 |
Table 6.
Accuracy of the PWV at the three meteorological stations.
Table 6.
Accuracy of the PWV at the three meteorological stations.
Meteorological Stations | Without Correction (mm) | With EMB, FTC and RRM Correction (mm) |
---|
RMS | MAE | MXAE | RMS | MAE | MXAE |
---|
54826 | 0.662 | 0.431 | 3.947 | 0.203 | 0.128 | 1.390 |
54836 | 0.221 | 0.094 | 2.532 | 0.097 | 0.048 | 1.153 |
54945 | 0.202 | 0.100 | 1.790 | 0.105 | 0.051 | 0.758 |