============================================================================== Dataset: CSU-DYNAMO upper-air and surface gridded analyses Version: 3b Release date: 29 September 2014 Corrected: 25 September 2015 (see Note #2 below) ============================================================================== *** NOTE *** ------------ (1) Some minor errors were detected in versions 2a and 2b (bad surface temperature at 4 times and bad surface water vapor mixing ratio at 03Z, 09Z, 15Z, and 21Z at several times in version 2b). These latter errors were seen primarily over the Southern Sounding Array (SSA). All of these errors have been corrected in versions 3a and 3b. (2) After the initial release of this version, a miscalculation was discovered in the derived fields (primarily with the vorticity field which was found to be twice a large as it should have been). A corrected version was released on Sept. 25, 2015. General Description: ------------------- Analyzed fields (see list below) were produced on 1 degree latitude/longitude grids, 25 hPa vertical resolution (1000 to 50 hPa) using the multiquadric interpolation scheme of Nuss and Titley (1994). Observations were supplemented with ECMWF operational analyses in data sparse regions. Analyses are available on the following grid and for the following times: ------------------------------------------------------------------------ * 20S - 20N, 35E - 155E * at 00, 03, 06, 09, 12, 15, 18 and 21 UTC * for 1 October thru 31 December 2011 * each data file covers 1 day Basic fields: ------------ z geopotential height (m) t temperature (C) q water vapor mixing ratio (g/kg) u zonal wind (m/s) v meridional wind (m/s) ps surface pressure (hPa) Derived fields: -------------- w vertical p-velocity (hPa/s) div horizontal divergence (1/s) vor horizontal vorticity (1/s) Q1 apparent heating (K/day) Q2 apparent drying (K/day) Notes: ----- (1) For upper-air fields, the first pressure level (labeled 1025 mb) contains the associated surface field. (2) A field is underground if the pressure level of the data is greater than the surface pressure (ps). (3) File format: netCDF self describing files See http://www.unidata.ucar.edu/software/netcdf/ for more info. (4) To quickly examine a datafile: ncdump netcdf_filename | more (5) Input dataset for gridded analyses: - This dataset version (3b) used ECMWF Operational Analyses (OA) in data sparse regions. ECMWF OA is used at 5 degree grid intersections if no observations were present within 4.5 degrees of such a grid intersection. No OA is used within core sounding arrays, however, OA is used (as described above) at 5 degree grid intersections near the arrays when one or both ships were offsite. - Hi-res (Level 4) sounding data were used at the following core sites (Male, Gan, Diego, Revelle, Mirai, Colombo) and 26 other Priority Sounding Sites (See Ciesielski et al. 2013). - Level 4 sounding data (uniform 5-hPa resolution with QC flags) are based on Level 3 (L3) sounding data (L3 is native resolution, humidity corrections applied as needed); see Ciesielski et al. (2013) for details of sounding dataset. - Hi-res P3 dropsonde data were used. - GTS-resolution (10-25hPa) sounding data were used at 54 other sites. - Soundings at Male' and Colombo were time interpolated to 3-hr resolution to match temporal resolution of sites in the Southern Sounding Array. - Because of the large station spacing between sites over the Northern Sounding Array, the strong diurnal cycle and flow-blocking at low-levels at Colombo are aliased onto the large-scale analyses and adversely affect the large-scale budgets. This version of the analyses used adjusted Colombo soundings which mute these effects. To mitigate these local island effects on the large-scale budgets, a procedure was designed which used low-level ECMWF-analyzed fields in the vicinity of Sri Lanka to estimate open-ocean conditions at Colombo’s location as if the island were not present. These “unperturbed” ECMWF fields at low-levels were then merged with the observed Colombo soundings. Details can be found in Ciesielski et al.(2014). - Sounding data were supplemented with CIMSS cloud drift winds, ASCAT surface winds, and COSMIC thermodynamic profiles (COSMIC moisture data were not used below 850 hPa due to issues with these data at lower-levels). - Johnson and Ciesielski (2013) describes the V1 gridded dataset (6hr, soundings only) and some preliminary findings. - Johnson et al. (2014, submitted to JAS) describes the V2 gridded dataset and atmospheric budgets. - If one wants to use the analyses with no model influence, then the V3a analyses should be used realizing that those analyses are most reliable when the ships are on-site. - The Revelle was onsite (nominal position 80.5E,Eq.) from 4 Oct. - 29 Oct., 10 Nov. - 04 Dec., and 18 Dec. - 31 Dec. - The Mirai was on site (nominal position 80.5E,8S) from 1 Oct. - 24 Oct. and 1 Nov. - 27 Nov. - Analyses over the NSA are less reliable after 5 Dec. when Colombo went from 4/day to 1/day sonde observations. - Male' ended sonde operation on 15 Dec. 2011 at 00Z so in effect there was no NSA after this date even though the Revelle returned on 18 Dec. - See the figure "invent_vis.png" which shows a visual inventory of soundings for 6 core Indian Ocean sites (Fig. 2 of Ciesielski et al. 2013). Contact information: ------------------- If any problems are noted in these datasets, please notify Paul Ciesielski (paulc@atmos.colostate.edu). References: ---------- Ciesielski, P. E. and coauthors, 2014: Quality-controlled upper-air sounding dataset for DYNAMO/CINDY/AMIE: Development and corrections. J. Atmos. Oceanic Technol., 31, 741-764. Ciesielski, P. E., R. H. Johnson, K. Yoneyama, and R. K. Taft, 2014: Mitigation of Sri Lanka island effects in Colombo sounding data and its impact on DYNAMO analyses. To appear in J. Meteor. Soc. Japan. Johnson, R. H., and P. E. Ciesielski, 2013: Structure and properties of Madden-Julian Oscillations deduced from DYNAMO sounding arrays. J. Atmos. Sci., 70, 3157-3179. Johnson, R. H. and coauthors, 2014: Sounding-Based Thermodynamic Budgets for DYNAMO. Submitted to J. Atmos. Sci. Nuss, W. A., and D. W. Titley, 1994: Use of multiquadric interpolation for meteorological objective analysis. Mon. Wea. Rev., 122, 1611-1631.