VS-Lite Model of Tree-Ring Width ----------------------------------------------------------------------- World Data Center for Paleoclimatology, Boulder and NOAA Paleoclimatology Program ----------------------------------------------------------------------- NOTE: PLEASE CITE ORIGINAL REFERENCE WHEN USING THIS MODEL!!!!! NAME OF DATA SET: VS-Lite Model of Tree-Ring Width LAST UPDATE: 10/2014 (debugging for using soil moisture inputs; bug fix for mistake in calculation of istar and I, following Huang et al (1996)) CONTRIBUTORS: Tolwinski-Ward, S.E., M.N. Evans, M.K. Hughes, and K.J. Anchukaitis. IGBP PAGES/WDCA CONTRIBUTION SERIES NUMBER: 2010-130 WDC PALEO CONTRIBUTION SERIES CITATION: Tolwinski-Ward, S.E., et al. 2010. VS-Lite Model of Tree-Ring Width. IGBP PAGES/World Data Center for Paleoclimatology Data Contribution Series # 2010-130. NOAA/NCDC Paleoclimatology Program, Boulder CO, USA. ORIGINAL REFERENCE: Tolwinski-Ward, S.E., M.N. Evans, M.K. Hughes, and K.J. Anchukaitis. 2011. An efficient forward model of the climate controls on interannual variation in tree-ring width. Climate Dynamics, 36: 11-12, 2419-2439, doi: 10.1007/s00382-010-0945-5 ABSTRACT: We present a simple, efficient, process-based forward model of tree-ring growth, called Vaganov-Shashkin-Lite (VS-Lite), that requires as inputs only latitude and monthly temperature and precipitation. Simulations of six bristlecone pine ring-width chronologies demonstrate the interpretability of model output as an accurate representation of the climatic controls on growth. Ensemble simulations by VS-Lite of two networks of North American ring-width chronologies correlate with observations at higher significance levels on average than simulations formed by regression of ring width on the principal components of the same monthly climate data. VS-Lite retains more skill outside of calibration intervals than does the principal components regression approach. It captures the dominant low- and high frequency spatiotemporal ring-width signals in the network with an inhomogeneous, multivariate relationship to climate. Because continuous meteorological data are most widely available at monthly temporal resolution, our model extends the set of sites at which forward-modeling studies are possible. ADDITIONAL REFERENCES: Tolwinski-Ward, S.E., K.J. Anchukaitis, and M.N. Evans. 2013. Bayesian parameter estimation and interpretation for an intermediate model of tree-ring width. Climate of the Past, 9,1-13, 2013 doi: 10.5194/cp-9-1-2013 Tolwinski-Ward, S.E., M.N. Evans, M.K. Hughes, and K.J. Anchukaitis. 2011. Erratum to: An efficient forward model of the climate controls on interannual variation in tree-ring width. Climate Dynamics, 36: 11012, 2441-2445, doi: 10.1007/s00382-011-1062-9 Daly, C., M. Halbleib, J.I. Smith, W.P. Gibson, M.K. Doggett, G.H. Taylor, J. Curtis, and P.A. Pasteris. 2008. Physiographically-sensitive mapping of temperature and precipitation across the conterminous United States. International Journal of Climatology, 28: 2031-2064 GEOGRAPHIC REGION: N/A PERIOD OF RECORD: N/A FUNDING SOURCES: US National Science Foundation NSF/CMG grant 0724802, and NSF DMS-1204892 US National Oceanic and Atmospheric Administration grants NOAA/CPO NA060AR4310115, NOAA NA07OAR4310060, and NOAA NA07OAR4310424. DESCRIPTION: Vaganov-Shashkin-Lite (VS-Lite) process-based forward model of tree-ring growth. Model Revision History v0.1 - Original coding at monthly timestep from full daily timestep model (SETW, 4/09) v1.0 - Changed soil moisture module to the CPC Leaky Bucket model (SETW, 5/09) v1.1 - No upper parametric bounds for gT, gW as in full model; no density module (SETW, 9/09) v1.2 - Added adjustable integration window parameters (SETW, 1/10) v2.0 - Minor debugging for Octave compatibility, final version for publication (SETW, 10/10) v2.1 - Error in evapotranspiration calculation corrected (SETW, 7/11) v2.2 - Option for soil moisture "substepping" added (SETW, 11/11) v2.3 - Add switch to allow moisture M to be given as input rather than estimated from T and P; add variable input options and improved commenting (SETW, 7/13) v2.4 - Error in using soil moisture inputs by switch 'hydroclim' corrected (MNE, 11/13) v2.5 - Corrected error in calculation of istar and I (MNE, 8/14). Thanks to Nick Graham (HRC) for discovering and pointing out the error. TEST RING WIDTH DATA are from the International Tree Ring Data Bank (http://www.ncdc.noaa.gov/paleo/treering.html) Site 1: 36.45N, -118.22E, 'ca530' Site 2: 34.17N, -117.12E, 'ca544' Site 3: 39.02N, -122.82E, 'ca615' Site 4: 39.32N, -106.08E, 'co523' TEST CLIMATE DATA are from OSU's PRISM climate product (http://www.prism.oregonstate.edu/) See Daly et al. 2008. If you find bugs or would like to suggest model extensions or improvements, please contact Suz Tolwinski-Ward, tolwinsk@ucar.edu --------------------------------------------------------------------- VSLite package contains the following files: 1. readme-vslite.txt (this file) 2. test_vslite_v2_5.m - test code loads test data, estimates parameters, runs model, plots output. 3. VSLite_v2_5.m - main model code. 4. VSLite_v2_2_documentation.pdf - documentation for updates in version 2.2 5. vslite_testdata.mat - file of test data 6. VSLite_correction_7_2011.pdf - documentation for correction to model code 7. possible-erratum-email.txt - regarding typographical error (fixed v2.1) in code 8. estimate_vslite_params_v2_3.m - parameter estimation code for VSLite v2.3 To test VS-lite code, type test_vslite_v2_5 at the MATLAB command prompt. Script is set up to load test data from vslite_testdata.mat, call parameter estimation code to estimate parameters (with default options), and to call the main code of VS-Lite to run test simulations (with all default options) and finally display output. Typing help VSLite_v2_5 and help estimate_vslite_params_v2_3 at the MATLAB command prompt will give more general instructions and list all user-options for running the VS-Lite model and the parameter estimation code. SETW 7/9/2013 (updated MNE 10/2014)