LDAS Phone
Conference Minutes
December 17th
2001
- a watershed phone conference, as Brian Cosgrove announced he had discovered another non-trivial error in his derivation of his 3- year retrospective LDAS forcing for the 3-year period of Oct 96 to Sep 99. The essence of the error was that the Stage-IV radar/gage hourly precipitation analysis was unknowingly and unintentionally ignored owing to a programming glitch and therefore the hourly disaggregation of the daily gage-only precipitation analysis was falling back to model-based 3-hourly EDAS precipitation fields and this yielded a 3-hourly persistence character in the resultant hourly precipitation forcing. Brian had executed reruns of the MOSAIC LSM with corrected forcing and he and Paul Houser concluded that the error had a non-trivial impact on MOSAIC soil moisture state variables, because MOSAIC of the method by which MOSAIC uses hourly precip variability to infer the fraction of total precip that is from convective sources.
- reluctantly, the group decided that the retrospective N-LDAS LSM runs had to be re-executed yet again, postponing the assembling of results for the upcoming group-authored BAMS paper. Some positive benefits of this re-execution is that the latest correction of the retrospective forcing will include also corrections to the small solar insolation error at low sun angles and will include more months in the 3-year period with a) GOES-based skin temperature retrievals for validation purposes and b) CAPE fields from EDAS for inference of fraction of convective precipitation.
- the groups reported on status of catch-up of realtime runs to the present time. Unlike situation at last phone conference (Nov 28), all four LSMs are now much closer to being caught up to present time. NOAH and MOSAIC are caught up to present time. SAC is caught up to Nov 12, and VIC is caught up to Sep 29.
- the group visited the Rutger's web site of Lifeng Luo and further examined and discussed his validation results of LDAS LSM and LDAS forcing against OU Mesonet observations, including soil moisture. His web site is viewable at http://climate.envsci.rutgers.edu/luo/research/LDAS/. His results to date show that both NOAH and VIC seem to underestimate (overestimate) the latent (sensible) heat flux in the warm season, whereas MOSAIC is just the opposite. Also, NOAH appears to overestimate the net longwave radiation.
- the remainder of the phone call focused on fleshing out further the strawman outline for the group-authored BAMS article that was presented at the Nov 14 meeting. Justin Sheffield of the VIC group and Lifeng Luo of Rutgers group will contribute validation results on solar insolation (to complement the main validation of Jesse Meng) and precipitation (to complement those of Lifeng Luo). A table summarizing features of the 4 LSMs will be included. Some time series plots of surface energy fluxes will be shown, both for individual days and monthly means (mainly from Lifeng Luo's validation results). The paper will include a "PILPS-like" scatter plot of 3-year balance of water (runoff versus evaporation) and energy (latent versus sensible). Additionally, the paper will include some of Jeff Basara's OU Mesonet variability of soil moisture and type around a single flux measuring site, to cast insight into some important variability issues to bear in mind when interpreting Lifeng Luo's soil moisture validation results. Brian Cosgrove volunteered to spearhead a section on spin-up behavior. John Schaake and Qingyun Duan of OHD and Eric Wood and Justin Sheffield will contribute to Dag's streamflow validation section. The OHD folks will also contribute to the monthly water balance material.
- Ken Mitchell will email out a) information of NCEP/NESDIS global monthly greenness fields to VIC group, b) Jesse Meng two web sites on retrospective and realtime solar insolation validation results for N-LDAS against SURFRAD, c) NESDIS contact for 4-year 1997-2001 archive of the NESDIS daily 4-km N.H. snow cover analysis.
- Ken Mitchell will email the two co-chief editors of BAMS and get their opinion on the suitability of our proposed group-authored LDAS article for BAMS. Qingyun Duan has already established an initial contact with them on this issue, but as an editor of BAMS, Qingyun feels their may be some conflict of issue in his pushing BAMS too hard on the BAMS LDAS article issue.