Updates and Evaluation of the PX-LSM in MM5 Jonathan E. Plcim* Atmospheric Sciences Modeling Division, NOAA, Research Triangle Park, NC 27711 pleim@hpcc.epa.gov Ph:919-541 -1336, fax:919-541 -1379 Aijun Xiu Environmental Programs, MCNC-North Carolina Supercomputing Center, Research Triangle Park, North Carolina 27709 1. INTRODUCTION Starting with Version 3.4 there is a new land surface model known as the Pleim-Xiu LSM available in the MM5 system. Pleim and Xiu (1995) described the initial development and testing of this land surface and PBL model for use in mesoscale models. Last year's workshop proceedings provided a basic description of the model and some evaluation (Pleim and Xiu, 2000). A recent journal article (Xiu and Pleim 2001) presents a more detailed description of the LSM and its implementation in MM5 and further evaluation. This paper outlines some bug fixes, updates, guidance for use, and more evaluation. 2. MODEL DESCRIPTION / The PX land surface model's key elements include soil moisture based on the Interactions between Soil, Biosphere, and Atmosphere (ISBA) model (Noilhan and Planton 1989), surface fluxes including parameterization of vegetation, and a non-local closure PBL mode! developed by Pleim and Chang (1992). The surface model includes a two-layer soil model with a 1- cm surface layer and a 1-m root zone layer. Evaporation has three pathways: direct soil surface evaporation, vegetative evapotranspiration, and evaporation from wet canopies. Ground surface (1 cm) temperature is computed from the surface energy balance using a force- restore algorithm for heat exchange within the soil, Stomatal conductance is parameterized according to root zone soil moisture, air temperature and air humidity, photosynthetically active radiation (PAR), and several vegetation parameters such as leaf area index (LAI) and minimum stomatal resistance. Although originally based on the ISBA model, the stomatal and canopy parameterizations are almost entirely new. New features include updated stomatal functions with respect to environmental parameters, and inclusion of a data assimilation scheme similar to the technique described by Bouttier et al. (1993). A simple parameterization for describing seasonal growth of vegetation, including leaf- out of deciduous trees, has also been developed and tested. Refer to Xiu and Pleim (2001) for the details of the model. 3. BUG FIXES In the initial release of MM5v3.4 there was a bug in the PX LSM such that any radiation scheme other than the surface-only option would not work correctly. This bug was fixed and the current code available from the NCAR ftp site (mesouser) is correct. A naming conflict between the PX LSM and the RRTM radiation scheme caused another problem. There was a subroutine named surface.F in /physics/pbl_sfc/pxpbl and also a common block named surface in /physics/radiation/rrtm/rrtm. F. The result was that the model would crash whenever both PX and RRTM were selected. We fixed this by renaming surface.F to surfpx.F. This fix is not yet in the mesouser code but will be for the next release. In the mean time, to run PX and RRTM together this change must be made. 4. UPDATES In addition to these bug fixes several updates have been made to the PX code which will be available in future releases. None of these modifications are necessary for use of the PX option. All of these updates are related to surface layer and PBL parameterizations and should be considered incremental R&D improvements. PBL Height The method for determination of the PBL height has been updated from the method described by " on assignment to the National Exposure Research Laboratory, US Environmental Protection Agency ------- Holtslag et al, (1990) to the slightly different method suggested by Holtslag et al. (1995). U. Estimation We have done two things to the IJ. calculation. We removed the minimum of 0.1 m/s and we removed the eonvective velocity from the wind speed used in the calculation of U. (see the MM5 model description: Grell et al. 1994). Figure 1 shows the effects of these changes. The curve labeled MM5 is the original version of MM5v3.4 run at 2 km grid resolution over central Tennessee and southern Kentucky. The curve labeled modified is a similar model run with the U. modifications. Note that the most pronounced effect of the eonvective velocity is the much higher [/.values in during the daytime. These U. values are quite unrealistic for these low wind speed conditions (see fig.2). stable regimes defined by Hogstrom as z/L greater than and less than 1, respectively. The details of these updates will be published in a technical note. An important effect of these changes is to greatly increase nocturnal heat flux compared to the original Blackadar formulations. Eddy diffusivity The current release of pxpbl includes the asymmetric eonvective model (ACM) which is a modification of the Blackadar (1978) non-local scheme for eonvective conditions. For non-convective conditions the eddy diffusivity is estimated according to Blackadar (1979) as in hirpbl. Since v3.4 we have updated the length scale such that it increases with heigth above ground with an asymptotic limit of 80 m. We also changed the Richardson number function to be U, obs U. MM5 U modified CC 0,3 Jul/4 Jul/5 Figure 1. Observed arid modeled U. at Keysburg, KY Jul/6 Flux-profile relationships Hie technique for estimating the _ functions from bulk Richardson number has been updated to conform more closely to the Hogstrom (1988) functions. The method in the current v3.4 release is essentially the same as in hirpbl (Blackadar 1979) using the 4 stability regimes. We have eliminated regime 3, which simply set the _ functions to zero, and have changed regimes 1 and 2 to reflect the stable and very similar to the form suggested by Liu and Carroll (1996). The net effect of these changes is to greatly decrease eddy diffusivity under stable conditions. Figure 2 shows the aggregate effect of these changes on wind speed at a site in southern Kentucky as simulated by MM5 at 2 km grid resolution. After all these updates the layer 1 wind speed (-19 m AGL) generally increases, especially at night. However, when adjusted to 2.5 m height using surface layer similarity ------- theory, the model results compare very favorably with moisture initialization is less important than for other the 2.5 m observations. LSMs. We have shown that when using the soil 6 t Obs (2.5 m) MM5 (layer 1) Modified (layer 1} —M—Modified (2,5 m) 5 4 TJ 3 a. 1 0 J— Jul/4 Jul/5 Jul/6 Figure2. Observed and modeled windspeed at Keysburg, KY. 5. USER GUIDANCE For detailed instructions on how to run the PX LSM the reader is referred to the NCAR MM5 website under the heading "What's New in Version 3 since its Release In this paper we offer additional guidance on soil moisture initialization, chaining together long series of runs, soil moisture data assimilation, and vegetation growth options. The two LSM options now available in the MM5 system as well as future additional LSMs have many similarities but also many important differences. The choice of model depends on the application since each model was developed for different purposes. The PX LSM was developed for mesoscale meteorology modeling used to drive atmospheric chemistry models for air quality research and policy. Therefore, the PX LSM was designed to be run primarily in retrospective mode for extended periods (weeks to months). Key parameters for air quality applications include PBL height, air temperature, boundary layer winds, and surface fluxes. Given our emphasis on long-term retrospective modeling, the indirect soil moisture data assimilation scheme and the vegetation growth algorithms are important features of the PX LSM. Consequently, soil moisture nudging scheme with very different soil moisture initializations the simulations converge in about 3-6 days (see Figure 3). Because the soil moisture assimilation scheme uses the surface analyses of temperature and humidity, the analysis FDDA options must be activated. 0 3 ....... ^ t-j ' in-Low init g ¦ ——.Hign init I o.28.: I 0.26.: \ "o ¦ E =5 0.24.1 m J 0.22.1 u £ ; Jun/16 Jun/17 Jun/18 Jun/19 Figure 3. Deep (lm) soil moisture modeled at Dickson, TN with 2 different initializations Typically we run 4-5 days then start the next run with a 2-12 hour overlap for spin-up. Soil moisture can be initialized in three different ways: 1. a -Low init -Hign init S. Jun/16 Jun/17 Jun/18 Jun/11 ------- generic initialization based on the moisture availability factor by land use, 2, interpolation from the Eta analyses of soil moisture, and 3. directly from MMINPUT. The first two options are used for the first of a series of runs. The third option is used to chain together runs into a continuous simulation. A utility program call Interppx is used to extract soil moisture and temperature for the initial time from a previous MMOUT file which are then added to MMINPUT. There are three vegetation growth options that can be important for extended runs to account for seasonal changes in vegetation parameters. The first option uses the monthly VEGFRAC fields provided by Terrain to scale LAI and vegetation fractions for each grid cell. The second option estimates LAI of natural vegetation according to the model's deep soil temperature and estimates LAI, roughness length, and vegetation fraction of agricultural vegetation using simple growth algorithms starting from crop emergence, which is estimated from gridded planting data information read in from an additional ASCII file. Note that this option was created for a specific project studying pesticides where the focus was on agricultural areas. The third option is similar to the second except that the planting dates are estimated from the VEGFRAC fields thus obviating the need for the additional planting date input file. The second and third options were specifically designed for the winter-spring- summer transition. These schemes are particularly well suited for simulating year-specific leaf-out and early season crop growth, but they are not recommended for the autumn season. Note that the PX LSM aggregates vegetation parameters by grid cell from fractional land use and soil data rather then using the dominant land use category. 6. FUTURE WORK Evaluation and development efforts are continuing. Our current focus is the simulation of the SOS Nashville 1999 field experiment. Ideas for future development include comparison of the current Jarvis-type stomatal conductance model with a photosynthesis based stomatal model. If the photosynthesis model proves superior, the current scheme will be replaced. Another goal is to decouple the LSM from the PBL model so that other PBL models could be used with the LSM. 7. ACKNOWLEDGMENT Our thanks to Tilden Meyers of NOAA/ARUATDD in Oak Ridge, TN for providing the measurement data from the Keysburg, KY site. 8. REFERENCES Blackadar, A. K., 1978: Modeling pollutant transfer during daytime convection. Preprints, Fourth Symp. on Atmospheric Turbulence, Diffusion, and Air Quality, Reno, NV, Amer. Meteor. Soc., 443-447. Blackadar, A.K., 1979: High resolution models of the planetary boundary layer. Advances in Environmental Science and Engineering, 1, Pfafflin and Ziegler, Eds., Gordon and Briech Sci. Publ., New York, 50-85. Bouttier, F., J. F. Mahfouf, and J. Noilhan, 1993: Sequential assimilation of soil moisture from atmospheric low- level parameters. Part I: Sensitivity and calibration studies. J. Appl. Meteor. 32, 1335-1351. Grell, G. A., J. Dudhia, and D. R. Stauffer, 1994, A Description of the Fifth-Generation Penn State/NCAR Mesoscale Model (MM5). NCAR Technical Note, NCAR/TN-398+STR, 122 pp. Hogstrom, U., 1988: Non-dimensional wind and temperature profiles in the atmospheric surface layer. Bound.-Layer Meteorol., 42, 55-78. Holtslag, A. A. M, de Bruijn, E. I. F., and Pan, H.-L., 1990: A high resolution air mass transformation model for short-range weather forecasting. Mon. Wea. Rev., 118, 1561-1575. Holtslag, A. A. M, E. V. Meijgaard, and W. C. DeRooy, 1995: A comparison of boundary layer diffusion schemes in unstable conditions over land. Bound. - Layer Meteorol., 76, 69-95. Liu, M. and J, I. Carroll, 1996: A high-resolution air pollution model suitable for dispersion studies in complex terrain. Mon. Wea. Rev., 124, 2396-2409. Noilhan, J., and S. Planton, 1989: A simple parameterization of land surface processes for meteorological models. Mon. Wea. Rev., 117, 536-549. Pleim, J. E., and J. S. Chang, 1992: A non-local closure model for vertical mixing in the convective boundary layer. Atmos. Environ., 26A, 965-981. Pleim, J. E„ and A. Xiu, 1995: Development and testing of a surface flux and planetary boundary layer model for application in mesoscale models. J. Appl Meteor,, 34, 16-32. Pleim, J. E, and A. Xiu, 2000: A new land-surface model in MM5. Preprints of The Tenth PSU/NCAR Mesoscale Mode! Users* Workshop, June 21-22, 2000, Boulder, CO Xiu, A., and J. E. Pleim, 2001: Development of a land surface model part I: Application in a mesoscale meteorology model. J. Appl. Meteor., 40,192-209. DISCLAIMER This paper has been reviewed in accordance with the US Environmental Protection Agency's peer and administrative review policies and approved for presentation and publication. ------- NERL-RTP-AMD-01-127 TECHNICAL REPORT DATA 1. REPORT NO. EPA/600/A-01/082 2. 3 4. TITLE AND SUBTITLE Updates and Evaluation of the PX-LSM in MM5 5.REPORT DATE 6.PERFORMING ORGANIZATION CODE 7. AUTHOR(S) Jonathan E. Pleim1 and Aijun XiuJ 8.PERFORMING ORGANIZATION REPORT NO. 9. PERFORMING ORGANIZATION NAME AND ADDRESS 'Same as Block 12 'Environmental Programs, MCNC, North Carolina Supercomputing Center, Research Triangle Park, NC 27709 10.PROGRAM ELEMENT NO. 11. CONTRACT/GRANT NO. 12. SPONSORING AGENCY NAME AND ADDRESS U.S. Environmental Protection Agency Office of Research and Development National Exposure Research Laboratory Research Triangle Park, NC 27711 13.TYPE OF REPORT AND PERIOD COVERED Proceedings, FY-01 14. SPONSORING AGENCY CODE , EPA/600/9 15. SUPPLEMENTARY NOTES 16. ABSTRACT Starting with Version 3.4 there is a new land surface model known as the Pleim-Xiu LSH available in the MM5 system. Pleim and Xiu (1995) described the initial development and testing of this land surface and workshop proceedings provided a basic description of the model and some evaluation (Pleim and Xiu, 2000). A recent journal article (Xiu and Pleim, 2001) presents a more detailed description of the LSM and its implementation in MM5 and further evaluation. This paper outlines some bug fixes, updates, guidance for use, and more evaluation. 17. KEY WORDS AND DOCUMENT ANALYSIS a. DESCRIPTORS b.IDENTIFIERS/ OPEN ENDED TERMS c.COSATI 18. DISTRIBUTION STATEMENT RELEASE TO PUBLIC 19. SECURITY CLASS (This Report) UNCLASSIFIED 21.NO. OF PAGES 4 20. SECURITY CLASS (This Page) UNCLASSIFIED 22. PRICE ------- |