Initiative of EUMETNET E-PROFILE and COST TOPROF for a ceilometer inter-comparison campaign

Transcription

1 European Ceilometer Network Initiative of EUMETNET E-PROFILE and COST TOPROF for a ceilometer inter-comparison campaign

2 The ceilometer inter-comparison campaign CeiLinEx2015 1) - Cloud detection and cloud base height - Ulrich Görsdorf - DWD, Meteorologisches Observatorium Lindenberg Ina Mattis - DWD, Meteorologisches Observatorium Hohenpeißenberg Gregor Pittke - Freie Universität Berlin and the CeiLinEx- Team (s. extended abstract) 1) Ceilometer Performance Experiment at Lindenberg 2015

3 Main goals of CeiLinEx2015: to study the systematic differences and uncertainties for the quantitative determination of attenuated backscatter profiles, to study the variations between systems of the same type and between different firmware-versions, respectively, to apply and to evaluate different calibration methods to investigate different algorithms for mixing layer heights to assess the cloud detection capability and to compare cloud base heights (CBH), especially for low clouds The last item is of great importance for aviation particularly in the context of automation. Results regarding the other goals are given in the poster of Margit Pattantyús-Abraham et al. P3(62) MOL-RAO

4 Experimental setup Experimental sites Time: June September 2015 Site: Richard-Aßmann Observatory Lindenberg Cloud Radar Raman lidar Total sky camera Radiosonde station 4 ascents/day Nubiscope OpenStreetMap Routine measuring program: Radiosoundings, eye observations, Raman-Lidar, cloud radar,. Human observations (24/7) MOL-RAO

5 Six different ceilometer types Campbell CS135 Vaisala CL51 Vaisala CL31 InGaAs 905 nm biaxial split-lense InGaAs 910 nm coaxial InGaAs 910 nm coaxial r = 0-10km r = 10 m t = 10 s r = 0-13km r = 10 m t = 15 s CL31 r = km r = 5 m t = 15 s Vaisala LD40 InGaAs 855 nm biaxial LD40 CL51 LUFFT CHM15k Nd:YAG 1064 nm biaxial CHM15kx LUFFT CHM15kx Nd:YAG 1064 nm biaxial r = 0-15 km r = 7.5 m t = 15 s r = 0-15 km r = 15 m t = 15 s r = 0-15 km r = 15 m t = 15 s MOL-RAO

6 MOL-RAO

7 Data management CHM15k FTP CHM15kx NetCDF NAS Archive NetCDF OPeNDAP CL31 NetCDF HTTP (Quicklooks) CL51 Conversion ASCII/binary QC/QA CS135 LD40 ASCII/binary Data acquisition host Published Data set (DOI ) MOL-RAO

8 Time series of CBH for Stratocumulus CBH 70 m MOL-RAO

9 Averaged profiles of scaled attenuated backscatter and CBH Height, km agl MOL-RAO

10 Time series of CBH for Cumulonimbus Rain rate > 10 mm/h mm/h MOL-RAO

11 Mean CBH differences related to the common mean value Deep clouds with rain CBH, m CBH, m CBH, m CBH, m MOL-RAO

12 Instrument-to-instrument variability (Scatter plots of cloud base heights, km) Stratocumulus System A System A System A System B CBH, km CBH, km CBH, km CBH, km System B CBH, km CBH, km CBH, km CBH, km MOL-RAO

13 Instrument-to-instrument variability (Scatter plots of cloud base heights, km) Stratus in rain System A System A System A System B CBH, km CBH, km CBH, km CBH, km System B CBH, km CBH, km CBH, km CBH, km MOL-RAO

14 Summary/Conclusions CeiLinEx2015 provided important insights into the performance of different ceilometer types. Differences of up to 70 m have been observed in base heights of low (liquid water) clouds. These are mainly caused by a lack of a quantitative definition of cloud base height. Need for a quantitative cloud base definition and standardized algorithm for cloud base determination. For precipitating deep clouds, differences of several kilometers were observed. More work is required to analyze the performance in cases where the cloud base is not well defined considering diverse technical concepts of the different ceilometer types (e.g. overlap effects). MOL-RAO

15 Additional slides MOL-RAO

16 Erroneously detected clouds MOL-RAO

17 Detection rate and Root Mean Square Differences CHM15k CHM15kx CL31 CL51 CS135 LD40 Sc St (rain) St (RR) Cum (dry) Nim (rain) DR RMSD DR RMSD DR RMSD DR RMSD DR RMSD MOL-RAO

18 Individual profiles of attenuated backscatter and CBH Height, km agl MOL-RAO

19 Time series of CBH for Cirrus MOL-RAO

20 Instrument-to-instrument variability (Root Mean Square Differences in m) A A A A A A B Stratocumulus Nimbostratus B B Stratus w/o rain Stratus with rain B MOL-RAO

21 Deviation of detection rates 1) to a common mean CHM15k CHM15kx CL31 CL51 CS135 LD40 Sc St St (rain) Cu (dry) Ns (rain) ) Detection rate = number of cloud hits / number of measurements MOL-RAO