17:15 〜 18:30
[ACG36-P08] Heavy orographic rainfall in Taiwan missed by the GPM Dual-Frequency Precipitation Radar (DPR) algorithm
キーワード：GPM/DPR、降水、地形性降水
1 Introduction
Because precipitation in orographic areas is difficult to observe with rain gauges or microwave radiometers, spaceborne precipitation radar observations are important. However, spaceborne precipitation radar observations also have a drawback: in orographic areas, shallow precipitation cannot be captured due to the strong influence of the ground surface clutter. The orographic areas of northern Taiwan are unique in the world in that they are equipped with a high density of rain gauges (Cheng and Yu 2019, J. Atmos. Sci.). In this study, we examined the spaceborne precipitation radar data by comparing with rain gauge data and ground-based radar data.
2 Data and Method
The Global Precipitation Measurement (GPM) mission’s Dual-frequency Precipitation Radar (DPR) was used as the spaceborne precipitation radar data. Although DPR contains two different frequencies, Ku-band radar (KuPR) is more sensitive to precipitation than Ka-band radar (KaPR). Therefore, KuPR Near Surface Rain (NSR) data was used in this study. NSR data is the rainfall intensity data at the lowest altitude that is not affected by surface clutter. The data was extracted from orbits passing through the analysis area from March 2014 to February 2020. For the rain gauge data, one-minute interval data from 22 Da-Tun rain gauge network (DTRGN) units installed in mountainous areas in northern Taiwan (Cheng and Yu 2019) were used. Data from the S-band ground-based Doppler radar on Wu-Fen-San (WFS) was also used to compare the vertical distribution of precipitation.
As a method to compare KuPR data and the rain gauge data, we compared a set of KuPR observation data within a radius of 3.5 km from each rain gauges, referring to Terao et al. (2017, SOLA).
3 Results
We found several cases which KuPR did not capture any rain at all, though rain gauges caught the rain. In particular, there were three cases which the rain gauges captured very strong rainfall of 10 mm/h or more, but the KuPR did not. The WFS radar also observed the rain for all the three cases.
The radar reflectivity profile of the WFS radar revealed that the heavy rainfall of 10mm/h or more is associated with shallow orographic rainfall systems below clutter free bottom (CFB) altitude estimated from KuPR. Although the effect of surface clutter is small near the nadir, one of the three cases was found to have a precipitation-top height lower than the near nadir CFB altitude (about 2300m).
Cheng and Yu (2019) showed that atmospheric stability tends to be higher in shallow precipitation in the orographic region of Taiwan. Also, Shige and Kummerow (2016, J. Atmos. Sci.) showed that there is a negative correlation between static stability and precipitation top height in Asian monsoon precipitation. The ERA5 reanalysis data revealed that the static stability around Taiwan became high. In these three cases, shallow precipitation was considered to have occurred more easily. In addition, in all three cases, moist air flowed into Taiwan from the cooler northern regions through the ocean, which might make rainfall in the orographic area.
Reducing the vertical extent of clutter region is important to capture the shallow rain lower than the CFB altitude. CFB altitude is estimated by using received power and could be estimated higher than it actually is (Iguchi et al. 2020, https://www.eorc.jaxa.jp/GPM/doc/algorithm/ATBD_DPR_202006_with_Appendix_a.pdf). The magnitude of the ground surface clutter is different between Ku-band and Ka-band. Therefore, the difference in radar reflectivity values of Ku-band and Ka-band, dual-frequency ratio (DFRm), jumps at the ground surface clutter. Using this fact, we examined that the CFB altitude could be set lower. As a result, we found that the DFRm value actually jumped at an altitude lower than the CFB altitude. By using this result, we may be able to broaden the clutter free region and capture the shallow rain at a lower altitude.
Because precipitation in orographic areas is difficult to observe with rain gauges or microwave radiometers, spaceborne precipitation radar observations are important. However, spaceborne precipitation radar observations also have a drawback: in orographic areas, shallow precipitation cannot be captured due to the strong influence of the ground surface clutter. The orographic areas of northern Taiwan are unique in the world in that they are equipped with a high density of rain gauges (Cheng and Yu 2019, J. Atmos. Sci.). In this study, we examined the spaceborne precipitation radar data by comparing with rain gauge data and ground-based radar data.
2 Data and Method
The Global Precipitation Measurement (GPM) mission’s Dual-frequency Precipitation Radar (DPR) was used as the spaceborne precipitation radar data. Although DPR contains two different frequencies, Ku-band radar (KuPR) is more sensitive to precipitation than Ka-band radar (KaPR). Therefore, KuPR Near Surface Rain (NSR) data was used in this study. NSR data is the rainfall intensity data at the lowest altitude that is not affected by surface clutter. The data was extracted from orbits passing through the analysis area from March 2014 to February 2020. For the rain gauge data, one-minute interval data from 22 Da-Tun rain gauge network (DTRGN) units installed in mountainous areas in northern Taiwan (Cheng and Yu 2019) were used. Data from the S-band ground-based Doppler radar on Wu-Fen-San (WFS) was also used to compare the vertical distribution of precipitation.
As a method to compare KuPR data and the rain gauge data, we compared a set of KuPR observation data within a radius of 3.5 km from each rain gauges, referring to Terao et al. (2017, SOLA).
3 Results
We found several cases which KuPR did not capture any rain at all, though rain gauges caught the rain. In particular, there were three cases which the rain gauges captured very strong rainfall of 10 mm/h or more, but the KuPR did not. The WFS radar also observed the rain for all the three cases.
The radar reflectivity profile of the WFS radar revealed that the heavy rainfall of 10mm/h or more is associated with shallow orographic rainfall systems below clutter free bottom (CFB) altitude estimated from KuPR. Although the effect of surface clutter is small near the nadir, one of the three cases was found to have a precipitation-top height lower than the near nadir CFB altitude (about 2300m).
Cheng and Yu (2019) showed that atmospheric stability tends to be higher in shallow precipitation in the orographic region of Taiwan. Also, Shige and Kummerow (2016, J. Atmos. Sci.) showed that there is a negative correlation between static stability and precipitation top height in Asian monsoon precipitation. The ERA5 reanalysis data revealed that the static stability around Taiwan became high. In these three cases, shallow precipitation was considered to have occurred more easily. In addition, in all three cases, moist air flowed into Taiwan from the cooler northern regions through the ocean, which might make rainfall in the orographic area.
Reducing the vertical extent of clutter region is important to capture the shallow rain lower than the CFB altitude. CFB altitude is estimated by using received power and could be estimated higher than it actually is (Iguchi et al. 2020, https://www.eorc.jaxa.jp/GPM/doc/algorithm/ATBD_DPR_202006_with_Appendix_a.pdf). The magnitude of the ground surface clutter is different between Ku-band and Ka-band. Therefore, the difference in radar reflectivity values of Ku-band and Ka-band, dual-frequency ratio (DFRm), jumps at the ground surface clutter. Using this fact, we examined that the CFB altitude could be set lower. As a result, we found that the DFRm value actually jumped at an altitude lower than the CFB altitude. By using this result, we may be able to broaden the clutter free region and capture the shallow rain at a lower altitude.