140 likes | 150 Views
This paper provides an overview of the QCRad algorithm for quality assessment and control of surface radiation measurements. It discusses the various tests and limits used to evaluate the data, including tests for clear-sky estimate, diffuse SW, direct component SW, comparison SW, upwelling SW, and up- and downwelling LW. The paper also explores additional tests that can be developed using measurements available at specific sites.
E N D
Overview of QCRad Irradiances Testing Chuck Long, CIRES/NOAA ESRL
Paper: • Long, C. N. and Y. Shi (2008): An Automated Quality Assessment and Control Algorithm for Surface Radiation Measurements, OASJ, 2, 23-37, doi: 10.2174/1874282300802010023.
QCRad • Enhanced from Long and Dutton Long CN and EG Dutton. 2002. “BSRN Global Network recommended QC tests, V2.0.” BSRN Technical Report. Available at http://ezksun3.ethz.ch/bsrn/admin/dokus/qualitycheck.pdf. • Uses limits and tests based on data from the particular site • Thus tailored to climate, etc. • Setting the limits assumes that most data used to determine limits are “good” • Classifies data as physically impossible, and 2 user defined limits
Global SW Tests Clear-Sky Estimate: SWclr = a x 0b a and b determined by user from plot inspection, typical values are 1050 Wm-2 and 1.2 • BSRN Physically Possible Min: -4 Wm-2 Max: Sa x 1.5 x 01.2 + 100 Wm-2 • BSRN Extremely Rare Min: -2 Wm-2 • User Configurable Max: Sa x D1 x 01.2 + 55 Wm-2 Max: Sa x C1 x 01.2 + 50 Wm-2
Diffuse SW Tests Tracker Off Test: • BSRN Physically Possible Min: -4 Wm-2 Max: Sa x 0.95 x 01.2 + 50 Wm-2 • BSRN Extremely Rare Min: -2 Wm-2 • User Configurable Max: Sa x D2 x 01.2 + 35 Wm-2Max: Sax C2 x 01.2 + 30 Wm-2 • Rayleigh Limit: RL = a0 + b02 + c03 + d04 + e05 + f0Prs
Direct Component SW Tests • BSRN Physically Possible Min: -4 Wm-2 Max: Sa x 0 • BSRN Extremely Rare Min: -2 Wm-2 • User Configurable Max: Sa x D3 x 01.2+ 15 Wm-2 Max: Sa x C3 x 01.2 + 10 Wm-2
Comparison SW Tests Non-Definitive Tests Black: IR Loss corrected Red: Not IR Loss corrected
Upwelling SW Tests • BSRN Physically Possible Min: -4 Wm-2 Max: Sa x 1.2 x 01.2 + 50 Wm-2 • BSRN Extremely Rare Min: -2 Wm-2 • User Configurable Max: Sa x D4 x 01.2 + 55 Wm-2 Max: Sa x C4 x 01.2 + 50 Wm-2 • Use air temperature to set limits: Must allow for snow covered ground if occurs at site
Up- and Downwelling LW Tests • BSRN Physically Possible Min: 40 Wm-2 Max Dn: 700 Wm-2 • Max Up: 900 Wm-2 • User Configurable LWdn Min: D5 Wm-2 (2nd level) Max: D6 Wm-2 (2nd level) Min: C5 Wm-2 (1st level) Max: C6 Wm-2 (1st level) LWup Min: D7 Wm-2 (2nd level) Max: D8 Wm-2 (2nd level) Min: C7 Wm-2 (1st level) Max: C8 Wm-2 (1st level)
Up- and Downwelling LW Tests Using Air Temperature • LWdn to Air Temperature comparison • D11 x Ta4 < LWdn < Ta4 + D12 • (2nd level) • C11 x Ta4 < LWdn < Ta4 + C12 • (1st level) • LWup to Air Temperature comparison • (Ta – D13 K)4 < LWup < (Ta + D14 K)4 • (2nd level) • (Ta – C13 K)4 < LWup < (Ta + C14 K)4 • (1st level)
LWdn versus LWup Comparison • LWdn to Lwup comparison • LWup - D15 Wm-2 <LWdn < LWup + D16 Wm-2(2nd level) • LWup - C15 Wm-2 <LWdn < LWup + C16 Wm-2(1st level)
Case & Dome vs Tair Temperatures • Ta - C17 < (Tc or Td) < Ta + C17 • (for both LWdn and LWup instruments. If have all 3, can determine "bad" one) • If Ta not available, test not possible. • C18 <= (Tc - Td) < C19 • If Tc and/or Td "bad", test not possible.
Other Tests • All Temperatures within climatological limits: • Tmin < Tx < Tmax • Where Tmin and Tmax are user defined minimum and maximum limits, respectively, determined using analysis of climatological data, and Tx is the temperature value being tested. Then if each Tc-Td pair is within +/- 10 K of each other, they are included in producing an average of all Tc and Td values that pass the testing. Ta must then fall within +/- 20 K of this average, and each individual Tc and Td must fall within +/- 15 K of this average, else the value is not used and is set to “-9999.0”. • Recap of other tests • Rayleigh atmosphere diffuse SW test • “Tracker Off” test • “Snow limit temperature” SWup test • Can we develop more tests using measurements available to us at the sites we’re interested in?
Next: • The QCRad FORTRAN Code Configuration settings file