{\rtf1\ansi\ansicpg1252\uc1 \deff0\deflang1033\deflangfe1033{\fonttbl{\f0\froman\fcharset0\fprq2{\*\panose 02020603050405020304}Times New Roman;}{\f1\fswiss\fcharset0\fprq2{\*\panose 020b0604020202020204}Arial;} {\f3\froman\fcharset2\fprq2{\*\panose 05050102010706020507}Symbol;}}{\colortbl;\red0\green0\blue0;\red0\green0\blue255;\red0\green255\blue255;\red0\green255\blue0;\red255\green0\blue255;\red255\green0\blue0;\red255\green255\blue0;\red255\green255\blue255; \red0\green0\blue128;\red0\green128\blue128;\red0\green128\blue0;\red128\green0\blue128;\red128\green0\blue0;\red128\green128\blue0;\red128\green128\blue128;\red192\green192\blue192;}{\stylesheet{\widctlpar\adjustright \fs20\cgrid \snext0 Normal;}{ \s1\keepn\widctlpar\adjustright \b\fs22\cgrid \sbasedon0 \snext0 heading 1;}{\*\cs10 \additive Default Paragraph Font;}{\s15\widctlpar\adjustright \f1\fs20\cgrid \snext15 HTML Body;}{\s16\widctlpar\adjustright \fs22\cgrid \sbasedon0 \snext16 Body Text;}} {\*\listtable{\list\listtemplateid67698689\listsimple{\listlevel\levelnfc23\leveljc0\levelfollow0\levelstartat1\levelspace0\levelindent0{\leveltext\'01\u-3913 ?;}{\levelnumbers;}\f3\fbias0 \fi-360\li360\jclisttab\tx360 }{\listname ;}\listid35858763} {\list\listtemplateid67698703\listsimple{\listlevel\levelnfc0\leveljc0\levelfollow0\levelstartat1\levelspace0\levelindent0{\leveltext\'02\'00.;}{\levelnumbers\'01;}\fbias0 \fi-360\li360\jclisttab\tx360 }{\listname ;}\listid249655411} {\list\listtemplateid67698703\listsimple{\listlevel\levelnfc0\leveljc0\levelfollow0\levelstartat1\levelold\levelspace0\levelindent360{\leveltext\'02\'00.;}{\levelnumbers\'01;}\fi-360\li360 }{\listname ;}\listid328024494}{\list\listtemplateid67698689 \listsimple{\listlevel\levelnfc23\leveljc0\levelfollow0\levelstartat1\levelspace0\levelindent0{\leveltext\'01\u-3913 ?;}{\levelnumbers;}\f3\fbias0 \fi-360\li360\jclisttab\tx360 }{\listname ;}\listid481000301}{\list\listtemplateid67698689\listsimple {\listlevel\levelnfc23\leveljc0\levelfollow0\levelstartat1\levelspace0\levelindent0{\leveltext\'01\u-3913 ?;}{\levelnumbers;}\f3\fbias0 \fi-360\li360\jclisttab\tx360 }{\listname ;}\listid714158312}{\list\listtemplateid67698703\listsimple{\listlevel \levelnfc0\leveljc0\levelfollow0\levelstartat1\levelold\levelspace0\levelindent360{\leveltext\'02\'00.;}{\levelnumbers\'01;}\fi-360\li360 }{\listname ;}\listid1209223736}{\list\listtemplateid67698703\listsimple{\listlevel\levelnfc0\leveljc0\levelfollow0 \levelstartat1\levelspace0\levelindent0{\leveltext\'02\'00.;}{\levelnumbers\'01;}\fi-360\li360\jclisttab\tx360 }{\listname ;}\listid1357853989}{\list\listtemplateid67698689\listsimple{\listlevel\levelnfc23\leveljc0\levelfollow0\levelstartat1\levelspace0 \levelindent0{\leveltext\'01\u-3913 ?;}{\levelnumbers;}\f3\fbias0 \fi-360\li360\jclisttab\tx360 }{\listname ;}\listid1433011237}{\list\listtemplateid67698703\listsimple{\listlevel\levelnfc0\leveljc0\levelfollow0\levelstartat1\levelold\levelspace0 \levelindent360{\leveltext\'02\'00.;}{\levelnumbers\'01;}\fi-360\li360 }{\listname ;}\listid1722748178}}{\*\listoverridetable{\listoverride\listid1209223736\listoverridecount0\ls1}{\listoverride\listid328024494\listoverridecount0\ls2} {\listoverride\listid1722748178\listoverridecount0\ls3}{\listoverride\listid249655411\listoverridecount0\ls4}{\listoverride\listid1357853989\listoverridecount0\ls5}{\listoverride\listid481000301\listoverridecount0\ls6}{\listoverride\listid714158312 \listoverridecount0\ls7}{\listoverride\listid35858763\listoverridecount0\ls8}{\listoverride\listid1433011237\listoverridecount0\ls9}}{\info{\title Ecosystem Model-Data Intercomparison Workshop}{\author Dick Olson}{\operator Tim Rhyne} {\creatim\yr2000\mo3\dy28\hr10\min21}{\revtim\yr2000\mo3\dy28\hr10\min21}{\printim\yr2000\mo3\dy15\hr22\min13}{\version2}{\edmins1}{\nofpages6}{\nofwords2299}{\nofchars13109}{\*\company }{\nofcharsws16098}{\vern89}} \widowctrl\ftnbj\aenddoc\formshade\viewkind4\viewscale90\pgbrdrhead\pgbrdrfoot \fet0\sectd \linex0\endnhere\sectdefaultcl {\*\pnseclvl1\pnucrm\pnstart1\pnindent720\pnhang{\pntxta .}}{\*\pnseclvl2\pnucltr\pnstart1\pnindent720\pnhang{\pntxta .}} {\*\pnseclvl3\pndec\pnstart1\pnindent720\pnhang{\pntxta .}}{\*\pnseclvl4\pnlcltr\pnstart1\pnindent720\pnhang{\pntxta )}}{\*\pnseclvl5\pndec\pnstart1\pnindent720\pnhang{\pntxtb (}{\pntxta )}}{\*\pnseclvl6\pnlcltr\pnstart1\pnindent720\pnhang{\pntxtb (} {\pntxta )}}{\*\pnseclvl7\pnlcrm\pnstart1\pnindent720\pnhang{\pntxtb (}{\pntxta )}}{\*\pnseclvl8\pnlcltr\pnstart1\pnindent720\pnhang{\pntxtb (}{\pntxta )}}{\*\pnseclvl9\pnlcrm\pnstart1\pnindent720\pnhang{\pntxtb (}{\pntxta )}}\pard\plain \s15\qc\widctlpar\adjustright \f1\fs20\cgrid {\b\f0\fs24 Ecosystem Model-Data Intercomparison Workshop \par Outlier Detection and Flagging for Class B Sites \par }{\f0\fs24 R.J. Olson, J.M.O. Scurlock, K.R. Johnson, and EMDI Workshop Participants \par March 15, 2000 \par }{\b\f0\fs22 \par }\pard \s15\widctlpar\adjustright {\b\f0\fs22 Summary \par {\pntext\pard\plain\s15 \f3\fs22\cgrid \loch\af3\dbch\af0\hich\f3 \'b7\tab}}\pard \s15\fi-360\li360\widctlpar\jclisttab\tx360{\*\pn \pnlvlblt\ilvl0\ls7\pnrnot0\pnf3\pnstart1\pnindent360\pnhang{\pntxtb \'b7}}\ls7\adjustright {\f0\fs22 Assigned consistent biome class to Class B sites (2363 records, 1271 sites) \par {\pntext\pard\plain\s15 \f3\fs22\cgrid \loch\af3\dbch\af0\hich\f3 \'b7\tab}}\pard \s15\fi-360\li360\widctlpar\jclisttab\tx360{\*\pn \pnlvlblt\ilvl0\ls7\pnrnot0\pnf3\pnstart1\pnindent360\pnhang{\pntxtb \'b7}}\ls7\adjustright {\f0\fs22 Calculated new NPP ensemble values (8 models) \par {\pntext\pard\plain\s15 \f3\fs22\cgrid \loch\af3\dbch\af0\hich\f3 \'b7\tab}}\pard \s15\fi-360\li360\widctlpar\jclisttab\tx360{\*\pn \pnlvlblt\ilvl0\ls7\pnrnot0\pnf3\pnstart1\pnindent360\pnhang{\pntxtb \'b7}}\ls7\adjustright {\f0\fs22 Calculated new AET ensemble values (4 models) \par {\pntext\pard\plain\s15 \f3\fs22\cgrid \loch\af3\dbch\af0\hich\f3 \'b7\tab}}\pard \s15\fi-360\li360\widctlpar\jclisttab\tx360{\*\pn \pnlvlblt\ilvl0\ls9\pnrnot0\pnf3\pnstart1\pnindent360\pnhang{\pntxtb \'b7}}\ls9\adjustright {\f0\fs22 Assigned unique Ids including MEAS_NUM for all NPP measurements (1-2363) and SITE_NUM for all unique sites by lat/long (1-1605). \par {\pntext\pard\plain\s15 \f3\fs22\cgrid \loch\af3\dbch\af0\hich\f3 \'b7\tab}}\pard \s15\fi-360\li360\widctlpar\jclisttab\tx360{\*\pn \pnlvlblt\ilvl0\ls9\pnrnot0\pnf3\pnstart1\pnindent360\pnhang{\pntxtb \'b7}}\ls9\adjustright {\f0\fs22 Rounded all lat/long coordinates to two decimal places (85 of the 1690 became non-unique) \par {\pntext\pard\plain\s15 \f3\fs22\cgrid \loch\af3\dbch\af0\hich\f3 \'b7\tab}}\pard \s15\fi-360\li360\widctlpar\jclisttab\tx360{\*\pn \pnlvlblt\ilvl0\ls9\pnrnot0\pnf3\pnstart1\pnindent360\pnhang{\pntxtb \'b7}}\ls9\adjustright {\f0\fs22 Reviewed biome type and flags for multiple measurements for multiple biomes at a site (399 records, dropped 187 because biome for a measurement was inconsistent with biomenew assigned to the site with multiple measurements, e.g., a site with several NPP measurements was assigned biomenew = \rquote Grassland C3\rquote but one of the measurements was for a \ldblquote Crop\rdblquote which was flagged.) \par {\pntext\pard\plain\s15 \f3\fs22\cgrid \loch\af3\dbch\af0\hich\f3 \'b7\tab}}\pard \s15\fi-360\li360\widctlpar\jclisttab\tx360{\*\pn \pnlvlblt\ilvl0\ls9\pnrnot0\pnf3\pnstart1\pnindent360\pnhang{\pntxtb \'b7}}\ls9\adjustright {\f0\fs22 Excluded 134 as managed sites (crops, pasture, plantation and wetlands) and 353 as outliers resulting in 1689 measurements \par {\pntext\pard\plain\s15 \f3\fs22\cgrid \loch\af3\dbch\af0\hich\f3 \'b7\tab}}\pard \s15\fi-360\li360\widctlpar\jclisttab\tx360{\*\pn \pnlvlblt\ilvl0\ls7\pnrnot0\pnf3\pnstart1\pnindent360\pnhang{\pntxtb \'b7}}\ls7\adjustright {\f0\fs22 Performed outlier analysis on NPP and driver data (20 flags) \par {\pntext\pard\plain\s15 \f3\fs22\cgrid \loch\af3\dbch\af0\hich\f3 \'b7\tab}}\pard \s15\fi-360\li360\widctlpar\jclisttab\tx360{\*\pn \pnlvlblt\ilvl0\ls7\pnrnot0\pnf3\pnstart1\pnindent360\pnhang{\pntxtb \'b7}}\ls7\adjustright {\f0\fs22 Identified a set of 353 sites to be excluded, including: \par }\pard \s15\li720\widctlpar\adjustright {\f0\fs22\ul One or more critical flags: \par }{\f0\fs22 164 flagged related to differences with NPP ensemble (bias, NE, MAE) \par 125 flagged related to high elevation (>2500 m) \par 35 flagged related to big differences between site and global precipitation \par 35 flagged related to big differences between site and global temperature \par }{\f0\fs22\ul Five or more less critical flags: \par }{\f0\fs22 Outside of .05 to .95 percentile by biome \par Outside of .95 CI determined by regressions with AET, PREC, TAVE \par {\pntext\pard\plain\s15 \f3\fs22\cgrid \loch\af3\dbch\af0\hich\f3 \'b7\tab}}\pard \s15\fi-360\li360\widctlpar\jclisttab\tx360{\*\pn \pnlvlblt\ilvl0\ls9\pnrnot0\pnf3\pnstart1\pnindent360\pnhang{\pntxtb \'b7}}\ls9\adjustright {\f0\fs22 Recalculated site means resulting in 933 sites with driver data and reasonable NPP. \par }\pard \s15\widctlpar\adjustright {\f0\fs22 \par }{\b\f0\fs22 Background - }{\f0\fs22 The Ecosystem Model-Data Intercomparison (EMDI) Workshop was held December 5-8, 1999 in Durham, New Hampshire with 12 modeling groups participating. The EMDI Workshop included a variety of models, including biogeochemical, satellite-driven, detailed process, and DVGM types. Initial results showed general agreement bet w een models and data but with obvious differences that indicate areas for potential data and model improvement. Much of the workshop was devoted to looking at potential outliers and harmonizing some of the driver data, especially land cover or vegetation types. \par \par }{\b\f0\fs22 Goal -}{\f0\fs22 The goal is to produce a consistent set of NPP measurements with associated environmental driver data that can be used for regional model development and validation. \par \par }\pard\plain \widctlpar\adjustright \fs20\cgrid {\b\fs22\cgrid0 What comprises an outlier? - }{\fs22\cgrid0 For the purposes of this discussion, an outl ier is defined initially as a data point considered to be unrepresentative of its location or land cover type, or otherwise \ldblquote difficult\rdblquote to represent in a generalized NPP model, such as took part in the EMDI exercise. However, there are also a number of re asons why an NPP data point may be designated an outlier, even when the data themselves are considered quite reasonable and representative of their location. \par \par }\pard\plain \s16\widctlpar\adjustright \fs22\cgrid {\cgrid0 Our analysis is based upon the complete package of NPP data and model drivers (climatology, soils , land cover type), not just the measured NPP data alone. Thus any problems experienced by the algorithms to produce consistent driver data (e.g. high elevation or highly incised topography, failure of the algorithms to represent the influence of the Gul f Stream on northern Scandinavia) may result in a point being labelled an outlier. Some characteristics peculiar to certain sites (e.g. a short growing season in \ldblquote cold desert\rdblquote grasslands) also resulted in outlier designation, since it may be hard to generalize the climate for such sites (e.g. mean annual temperature would appear too low in relation to NPP). \par }\pard\plain \widctlpar\adjustright \fs20\cgrid {\fs22\cgrid0 \par The process of synthesis, review and generalization of data sets may also have resulted in some points being designated outliers due to differences in methodologies for estimating NPP. For this first phase of EMDI, the large number of available NPP data allowed the designation of outliers to be fairly liberal. Future model-data intercomparison may require a more selective and conservatively justified elimination of outliers, in order to provide a more stringent test of the models. \par \par }{\b\fs22 QA Issues -}{\fs22 The ORNL group had reviewed the data extensively prior to EMDI by looking at scatter plots and data outside of reasonable limits. However, the Workshop provided the initial model results to compare with the observed NPP data. An ensemble NPP value was calculated for each site as the average of the 12 Class A (old class 1-2) and 8 Class B (old Class 3) models, including AVIM, CARAIB, CENTURY, GLO-PEM, IBIS, PnET, STOMATE, and VECODE. In addition, an ensemble AET (actual evapotranspiration) value was calculated based on the average of AET provided by four of the models. The EMDI workshop also provided an opportunity to review the classification of the sites, comp are observed to predicted NPP, and look at relationships between variables. The specific issues that we addressed are: \par }\pard\plain \s15\widctlpar\adjustright \f1\fs20\cgrid {\f0\fs22 \par {\pntext\pard\plain\s15 \f3\fs22\cgrid \loch\af3\dbch\af0\hich\f3 \'b7\tab}}\pard \s15\fi-360\li360\widctlpar\jclisttab\tx360{\*\pn \pnlvlblt\ilvl0\ls6\pnrnot0\pnf3\pnstart1\pnindent360\pnhang{\pntxtb \'b7}}\ls6\adjustright {\f0\fs22 Land cover / Biome Class Consistency- This review was prompted in part by the realization of problems in using the satellite-derived lan d cover for each site (i.e., often this represented the dominant land cover for a 1x1 km area, not the typical 1x1 m to 1 hectare NPP measurement site). \par {\pntext\pard\plain\s15 \f3\fs22\cgrid \loch\af3\dbch\af0\hich\f3 \'b7\tab}}\pard \s15\fi-360\li360\widctlpar\jclisttab\tx360{\*\pn \pnlvlblt\ilvl0\ls6\pnrnot0\pnf3\pnstart1\pnindent360\pnhang{\pntxtb \'b7}}\ls6\adjustright {\f0\fs22 Managed sites - In addition, modelers decided to flag and exclude likely heavily managed sites and wetlands from the EMDI comparison. \par {\pntext\pard\plain\s15 \f3\fs22\cgrid \loch\af3\dbch\af0\hich\f3 \'b7\tab}}\pard \s15\fi-360\li360\widctlpar\jclisttab\tx360{\*\pn \pnlvlblt\ilvl0\ls6\pnrnot0\pnf3\pnstart1\pnindent360\pnhang{\pntxtb \'b7}}\ls6\adjustright {\f0\fs22 Multiple NPP values for a site \endash Some sites have up to 35 observed NPP values, often from several vegetation types. We assume this is often a result of reporting imprecise latitude/longitude coordinates. Each site was a ssigned a biome and the biome of every NPP measurements at the site review with those inconsistent with the site biome were flagged. This resulted in 187 measurements being dropped. \par {\pntext\pard\plain\s15 \f3\fs22\cgrid \loch\af3\dbch\af0\hich\f3 \'b7\tab}}\pard \s15\fi-360\li360\widctlpar\jclisttab\tx360{\*\pn \pnlvlblt\ilvl0\ls6\pnrnot0\pnf3\pnstart1\pnindent360\pnhang{\pntxtb \'b7}}\ls6\adjustright {\f0\fs22 Coordinates \endash Coordinates for sites are reported in a mix of formats, from whole degrees for a set of Russian sites to 4 decimal places for GPS registered sites. Often the conversion from degrees-minutes appeared to have more precision than was the case, e.g., 80}{ \f0\fs22 {\field{\*\fldinst SYMBOL 176 \\f "Symbol" \\s 11}{\fldrslt\f3\fs22}}}{\f0\fs22 20\rquote = 80.3333}{\f0\fs22 {\field{\*\fldinst SYMBOL 176 \\f "Symbol" \\s 11}{\fldrslt\f3\fs22}}}{\f0\fs22 . Some of the climate data and model outputs were provided with coordinates rounded off to 2 decimal places. Although we had distributed data with coordinates up to 4 decimal places, we rounded all sites to 2 decimal places for consistency. \par }\pard \s15\widctlpar\adjustright {\f0\fs22 \par The power of the statistical-empirical approach is that we can look for patterns within similar groups (i.e., biomes) and look for relationships between variables. Even if we could review the original literature reference associated with each study, we would not pickup the potential outliers that were found in this process. \par \par }\pard\plain \s16\widctlpar\adjustright \fs22\cgrid {\b Biomes - }{ The outlier process started by reviewing the biome designation for all sites. Twenty-one classes were defined at the EMDI workshop to represent the data and needs of the models. They were assigned based on initial biome class, subbiome , species, vegetation type and evaluation by Jonathan Scurlock, Peter Thornton, Mac Post, Bill Parton, Steve DeGrosa and others. Four classes were assigned to heavily managed or sites typically not addressed by regional models. The types included crops, pasture, plantations, and wetlands. These sites will be flagged and generally excluded from the EMDI exercises. The 21 biome classes were grouped into 12 classes based on the number of sites to ensure there were enough data (at least 30-40 sites) within each biome to conduct the outlier detection described below. \par }\pard\plain \widctlpar\adjustright \fs20\cgrid {\fs22 \par }{\b\fs22 Biomes: \par }{\fs22 \par }\trowd \trgaph108\trleft-108\trbrdrt\brdrs\brdrw10 \trbrdrl\brdrs\brdrw10 \trbrdrb\brdrs\brdrw10 \trbrdrr\brdrs\brdrw10 \trbrdrh\brdrs\brdrw10 \trbrdrv\brdrs\brdrw10 \clvertalt\clbrdrt\brdrs\brdrw10 \clbrdrl\brdrs\brdrw10 \clbrdrb\brdrs\brdrw10 \clbrdrr \brdrs\brdrw10 \cltxlrtb \cellx2610\clvertalt\clbrdrt\brdrs\brdrw10 \clbrdrl\brdrs\brdrw10 \clbrdrb\brdrs\brdrw10 \clbrdrr\brdrs\brdrw10 \cltxlrtb \cellx3780\clvertalt\clbrdrt\brdrs\brdrw10 \clbrdrl\brdrs\brdrw10 \clbrdrb\brdrs\brdrw10 \clbrdrr \brdrs\brdrw10 \cltxlrtb \cellx6480\pard \widctlpar\intbl\adjustright {\fs22 BIOMENEW\cell Total\cell }{\b\fs22 Aggregated (}{\fs22 BIOME2)\cell }\pard \widctlpar\intbl\adjustright {\fs22 \row }\trowd \trgaph108\trleft-108\trbrdrt\brdrs\brdrw10 \trbrdrl \brdrs\brdrw10 \trbrdrb\brdrs\brdrw10 \trbrdrr\brdrs\brdrw10 \trbrdrh\brdrs\brdrw10 \trbrdrv\brdrs\brdrw10 \clvertalt\clbrdrt\brdrs\brdrw10 \clbrdrl\brdrs\brdrw10 \clbrdrb\brdrs\brdrw10 \clbrdrr\brdrs\brdrw10 \cltxlrtb \cellx2610\clvertalt\clbrdrt \brdrs\brdrw10 \clbrdrl\brdrs\brdrw10 \clbrdrb\brdrs\brdrw10 \clbrdrr\brdrs\brdrw10 \cltxlrtb \cellx3780\clvertalt\clbrdrt\brdrs\brdrw10 \clbrdrl\brdrs\brdrw10 \clbrdrb\brdrs\brdrw10 \clbrdrr\brdrs\brdrw10 \cltxlrtb \cellx6480\pard \widctlpar\intbl\adjustright {\fs22 *crops\cell 14\cell managed\cell }\pard \widctlpar\intbl\adjustright {\fs22 \row }\pard \widctlpar\intbl\adjustright {\fs22 *pasture\cell 17\cell managed\cell }\pard \widctlpar\intbl\adjustright {\fs22 \row }\pard \widctlpar\intbl\adjustright {\fs22 *plantation\cell 27\cell managed\cell }\pard \widctlpar\intbl\adjustright {\fs22 \row }\pard \widctlpar\intbl\adjustright {\fs22 *wetland\cell 46\cell managed\cell }\pard \widctlpar\intbl\adjustright {\fs22 \row }\pard \widctlpar\intbl\adjustright {\fs22 DBL forest / boreal\cell 43\cell boreal\cell }\pard \widctlpar\intbl\adjustright {\fs22 \row }\pard \widctlpar\intbl\adjustright {\fs22 DBL forest / temperate\cell 233\cell DBL forest / temperate\cell }\pard \widctlpar\intbl\adjustright {\fs22 \row }\pard \widctlpar\intbl\adjustright {\fs22 DBL forest / tropical\cell 17\cell DBL forest / tropical\cell }\pard \widctlpar\intbl\adjustright {\fs22 \row }\pard \widctlpar\intbl\adjustright {\fs22 Desert\cell 26 \cell desert\cell }\pard \widctlpar\intbl\adjustright {\fs22 \row }\pard \widctlpar\intbl\adjustright {\fs22 DNL forest / boreal\cell 29\cell boreal\cell }\pard \widctlpar\intbl\adjustright {\fs22 \row }\pard \widctlpar\intbl\adjustright {\fs22 EBL forest / temperate\cell 250\cell EBL forest / temperate\cell }\pard \widctlpar\intbl\adjustright {\fs22 \row }\pard \widctlpar\intbl\adjustright {\fs22 EBL forest / tropical\cell 102\cell EBL forest / tropical\cell }\pard \widctlpar\intbl\adjustright {\fs22 \row }\pard \widctlpar\intbl\adjustright {\fs22 ENL forest / boreal\cell 117\cell ENL forest / boreal\cell }\pard \widctlpar\intbl\adjustright {\fs22 \row }\pard \widctlpar\intbl\adjustright {\fs22 ENL forest / temperate\cell 210\cell ENL forest / temperate\cell }\pard \widctlpar\intbl\adjustright {\fs22 \row }\pard \widctlpar\intbl\adjustright {\fs22 grassland / C3\cell 41\cell grassland \cell }\pard \widctlpar\intbl\adjustright {\fs22 \row }\pard \widctlpar\intbl\adjustright {\fs22 grassland / C4 temperate\cell 18\cell grassland \cell }\pard \widctlpar\intbl\adjustright {\fs22 \row }\pard \widctlpar\intbl\adjustright {\fs22 grassland / C4 tropical\cell 32\cell grassland \cell }\pard \widctlpar\intbl\adjustright {\fs22 \row }\pard \widctlpar\intbl\adjustright {\fs22 Mediterranean\cell 12\cell Savanna\cell }\pard \widctlpar\intbl\adjustright {\fs22 \row }\pard \widctlpar\intbl\adjustright {\fs22 mixed forest\cell 49\cell mixed forest\cell }\pard \widctlpar\intbl\adjustright {\fs22 \row }\pard \widctlpar\intbl\adjustright {\fs22 Savanna / temperate\cell 1\cell Savanna \cell }\pard \widctlpar\intbl\adjustright {\fs22 \row }\pard \widctlpar\intbl\adjustright {\fs22 Savanna / tropical\cell 8\cell Savanna \cell }\pard \widctlpar\intbl\adjustright {\fs22 \row }\pard \widctlpar\intbl\adjustright {\fs22 Tundra\cell 24\cell tundra\cell }\pard \widctlpar\intbl\adjustright {\fs22 \row }\trowd \trgaph108\trleft-108\trbrdrt\brdrs\brdrw10 \trbrdrl\brdrs\brdrw10 \trbrdrb \brdrs\brdrw10 \trbrdrr\brdrs\brdrw10 \trbrdrh\brdrs\brdrw10 \trbrdrv\brdrs\brdrw10 \clvertalt\clbrdrt\brdrs\brdrw10 \clbrdrl\brdrs\brdrw10 \clbrdrb\brdrs\brdrw10 \clbrdrr\brdrs\brdrw10 \cltxlrtb \cellx2610\clvertalt\clbrdrt\brdrs\brdrw10 \clbrdrl \brdrs\brdrw10 \clbrdrb\brdrs\brdrw10 \clbrdrr\brdrs\brdrw10 \cltxlrtb \cellx3780\clvertalt\clbrdrt\brdrs\brdrw10 \clbrdrl\brdrs\brdrw10 \clbrdrb\brdrs\brdrw10 \clbrdrr\brdrs\brdrw10 \cltxlrtb \cellx6480\pard \widctlpar\intbl\adjustright {\fs22 Grand Total\cell 1317\cell \cell }\pard \widctlpar\intbl\adjustright {\fs22 \row }\pard \widctlpar\adjustright {\fs22 \par }\pard\plain \s1\keepn\widctlpar\outlinelevel0\adjustright \b\fs22\cgrid {Classes \par }\pard\plain \widctlpar\adjustright \fs20\cgrid {\fs22 \par At the EMDI Workshop, we decided to rename the groups to Class A (old Class 1 &2), Class B (old Class 3) and Class C (old Regional cells. This analysis covers Class B sites. A similar approach will be used with Class A and Class C sites. \par \par }\pard\plain \s1\keepn\widctlpar\outlinelevel0\adjustright \b\fs22\cgrid {Class B Sites \par }\pard\plain \widctlpar\adjustright \fs20\cgrid {\fs22 \par The EMDI NPP and driver data were reviewed to flag potential outliers based on criteria to identify unrepresentative sites or potential errors. Each variable was reviewed indepen dently and then in combination with other variables. At the EMDI Workshop, the QA checking was restricted to average NPP at 918 Class B (old Class 3) sites, those for which a complete set of model predictions from 7 models were available. In this phase, we reviewed all 2363 individual NPP data values that had model predictions from at least 3 models. The final step will be to calculate average NPP for all unique site-biome combinations using the subset of NPP records that pass our flagging criteria. In itially there were 1690 Class B sites; however, this set was reduced to 1271 unique sites with valid driver data. We expect after the exclusion of outliers that there will be approximately 1000 sites remaining. \par \par }\pard\plain \s16\widctlpar\adjustright \fs22\cgrid {The approach included the following tests to set flags: \par }\pard\plain \widctlpar\adjustright \fs20\cgrid {\fs22 \par {\pntext\pard\plain\fs22\cgrid \hich\af0\dbch\af0\loch\f0 1.\tab}}\pard \fi-360\li360\widctlpar\jclisttab\tx360{\*\pn \pnlvlbody\ilvl0\ls5\pnrnot0\pndec\pnstart1\pnindent360\pnhang{\pntxta .}}\ls5\adjustright {\fs22 Flags based on values outside of reasonable limits: \par }\pard \li720\widctlpar{\*\pn \pnlvlcont\ilvl0\ls0\pnrnot0\pndec }\adjustright {\fs22 \par ANPP_C>2000 gC/m2, BNPP_C>2000, TNPP_C > 3000 \par ELEV > 2500 m (high elevation sites were expected to comprise unrepresentative sub-biomes or present problems with the climate extrapolation algorithm) \par ANPP > .95TNPP, BNPP>.95 TNPP (for a site with both above and below ground components) \par }\pard \widctlpar{\*\pn \pnlvlcont\ilvl0\ls0\pnrnot0\pndec }\adjustright {\fs22 \par {\pntext\pard\plain\fs22\cgrid \hich\af0\dbch\af0\loch\f0 2.\tab}}\pard \fi-360\li360\widctlpar\jclisttab\tx360{\*\pn \pnlvlbody\ilvl0\ls5\pnrnot0\pndec\pnstart1\pnindent360\pnhang{\pntxta .}}\ls5\adjustright {\fs22 Flags based on questionable values. }{ \i\fs22 }{\fs22 \par }\pard \widctlpar{\*\pn \pnlvlcont\ilvl0\ls0\pnrnot0\pndec }\adjustright {\fs22 \par }\pard \li720\widctlpar{\*\pn \pnlvlcont\ilvl0\ls0\pnrnot0\pndec }\adjustright {\fs22 LAT_DD, LONG_DD \par }\pard \widctlpar{\*\pn \pnlvlcont\ilvl0\ls0\pnrnot0\pndec }\adjustright {\fs22 \par {\pntext\pard\plain\fs22\cgrid \hich\af0\dbch\af0\loch\f0 3.\tab}}\pard \fi-360\li360\widctlpar\jclisttab\tx360{\*\pn \pnlvlbody\ilvl0\ls5\pnrnot0\pndec\pnstart1\pnindent360\pnhang{\pntxta .}}\ls5\adjustright {\fs22 Flags based on NPP values outside of the .05-.95 percentiles for each biome calculated assuming a normal distribution of variables (this replaces our initial rule of using 2 standard deviations): \par }\pard \widctlpar{\*\pn \pnlvlcont\ilvl0\ls0\pnrnot0\pndec }\adjustright {\fs22 \par }\pard \li720\widctlpar{\*\pn \pnlvlcont\ilvl0\ls0\pnrnot0\pndec }\adjustright {\fs22 ANPP, BNPP, TNPP \par }\pard \widctlpar{\*\pn \pnlvlcont\ilvl0\ls0\pnrnot0\pndec }\adjustright {\fs22 \par {\pntext\pard\plain\fs22\cgrid \hich\af0\dbch\af0\loch\f0 4.\tab}}\pard \fi-360\li360\widctlpar\jclisttab\tx360{\*\pn \pnlvlbody\ilvl0\ls5\pnrnot0\pndec\pnstart1\pnindent360\pnhang{\pntxta .}}\ls5\adjustright {\fs22 Flags based on c limate values outside of the .01-.99 percentiles for each biome calculated assuming a normal distribution of variables. In addition, we set some specific limits for some biomes: \par }\pard \widctlpar{\*\pn \pnlvlcont\ilvl0\ls0\pnrnot0\pndec }\adjustright {\fs22 \par }\pard \li720\widctlpar{\*\pn \pnlvlcont\ilvl0\ls0\pnrnot0\pndec }\adjustright {\fs22 Temp, Precip \par Temp > 6 C for boreal forests \par Precip > 1000 mm for desert and tundra \par Precip < 1000 mm for tropical forest \par }\pard \widctlpar{\*\pn \pnlvlcont\ilvl0\ls0\pnrnot0\pndec }\adjustright {\fs22 \par {\pntext\pard\plain\fs22\cgrid \hich\af0\dbch\af0\loch\f0 5.\tab}}\pard \fi-360\li360\widctlpar\jclisttab\tx360{\*\pn \pnlvlbody\ilvl0\ls5\pnrnot0\pndec\pnstart1\pnindent360\pnhang{\pntxta .}}\ls5\adjustright {\fs22 Flags based on inconsistencies at a site, such as precipitation reported for the site being different than the precipitation derived from global climate data. These flags were based on calculating the Normalize d Error (NE) as the ratio of the difference (predicted - observed) divided by the average (predicted + observed )/2. Based on the frequency distribution of the ratios, ratios greater than 1.0 were flagged. \par }\pard \widctlpar{\*\pn \pnlvlcont\ilvl0\ls0\pnrnot0\pndec }\adjustright {\fs22 \par }\pard \li720\widctlpar{\*\pn \pnlvlcont\ilvl0\ls0\pnrnot0\pndec }\adjustright {\fs22 Elevation, Precipitation, Temperature \par }\pard \widctlpar{\*\pn \pnlvlcont\ilvl0\ls0\pnrnot0\pndec }\adjustright {\fs22 \par {\pntext\pard\plain\fs22\cgrid \hich\af0\dbch\af0\loch\f0 6.\tab}}\pard \fi-360\li360\widctlpar\jclisttab\tx360{\*\pn \pnlvlbody\ilvl0\ls5\pnrnot0\pndec\pnstart1\pnindent360\pnhang{\pntxta .}}\ls5\adjustright {\fs22 Flags ba sed on the comparison of measured NPP versus modeled NPP using the average or ensemble value for all available models. The comparison was based on bias (predicted - observed), Normalized Error (predicted - observed) divided by the average (predicted - ob served )/2, and Mean Absolute Error (MAE) \endash (predicted - observed) divided by the observed. . Based on the frequency distributions, bias greater than +_ 1000 gC/m2, NE ratios greater than +_1.0, and MAE ratios greater than +_ 5 were flagged. \par }\pard \widctlpar{\*\pn \pnlvlcont\ilvl0\ls0\pnrnot0\pndec }\adjustright {\fs22 \par }\pard \li720\widctlpar{\*\pn \pnlvlcont\ilvl0\ls0\pnrnot0\pndec }\adjustright {\fs22 NPP vs MODEL ensemble \par }\pard \widctlpar{\*\pn \pnlvlcont\ilvl0\ls0\pnrnot0\pndec }\adjustright {\fs22 \par {\pntext\pard\plain\fs22\cgrid \hich\af0\dbch\af0\loch\f0 7.\tab}}\pard \fi-360\li360\widctlpar\jclisttab\tx360{\*\pn \pnlvlbody\ilvl0\ls5\pnrnot0\pndec\pnstart1\pnindent360\pnhang{\pntxta .}}\ls5\adjustright {\fs22 Flags based on relationships between variables. Linear regression analysis was performed between NPP and average AET (from 4 models), NPP and precipitation, and NPP and temperature. Points falling outside of the .95 Confidence Interval about th e regression line were flagged. \par }\pard \widctlpar{\*\pn \pnlvlcont\ilvl0\ls0\pnrnot0\pndec }\adjustright {\fs22 \par }\pard \li720\widctlpar{\*\pn \pnlvlcont\ilvl0\ls0\pnrnot0\pndec }\adjustright {\fs22 NPP vs AET, NPP vs Prec, NPP vs Tave \par }\pard \widctlpar{\*\pn \pnlvlcont\ilvl0\ls0\pnrnot0\pndec }\adjustright {\fs22 \par {\pntext\pard\plain\fs22\cgrid \hich\af0\dbch\af0\loch\f0 8.\tab}}\pard \fi-360\li360\widctlpar\jclisttab\tx360{\*\pn \pnlvlbody\ilvl0\ls5\pnrnot0\pndec\pnstart1\pnindent360\pnhang{\pntxta .}}\ls5\adjustright {\fs22 The data were review at the EMDI workshop based on visual inspection and deviations from the model ensemble values. Those sites identified as outliers at EMDI were assigned flags in this analysis. \par }\pard \widctlpar\adjustright {\fs22 \par \par }\pard\plain \s1\keepn\widctlpar\outlinelevel0\adjustright \b\fs22\cgrid {Critical flags \par }\pard\plain \widctlpar\adjustright \fs20\cgrid {\fs22 \par We assume that sites that have multiple flags have inconsistencies between NPP, driver data, and model predictions, and are therefore more likely to be an outlier. This seemed to hold true in plots (see below). We also know that some flags are deserve more critical weight as indicators of potential problems. We designated flags by assigning a value of 10 for those critical checks, including NPP inconsistent with model NPP ensemble value, elevation > 2500 m or < -100 m, and either site temperature or precipitation inconsistent with that assigned based on the global climate data. We assigned a flag of 100 to the heavily managed sites for easy identification. \par \par The overall flags value was calculated as the sum of the individual flags. Sites with a sum greater that 100 (managed biomes) were dropped, and those greater than 5 (all those with at least one major flag) were considered potential for excluding. \par \par {\pntext\pard\plain\fs22\cgrid \hich\af0\dbch\af0\loch\f0 1.\tab}}\pard \fi-360\li720\widctlpar\jclisttab\tx720{\*\pn \pnlvlbody\ilvl0\ls4\pnrnot0\pndec\pnstart1\pnindent360\pnhang{\pntxta .}}\ls4\adjustright {\fs22 Drop managed sites (134 or 6% of the total) \par {\pntext\pard\plain\fs22\cgrid \hich\af0\dbch\af0\loch\f0 2.\tab}}\pard \fi-360\li720\widctlpar\jclisttab\tx720{\*\pn \pnlvlbody\ilvl0\ls4\pnrnot0\pndec\pnstart1\pnindent360\pnhang{\pntxta .}}\ls4\adjustright {\fs22 Select sites with no flags s et or less than 5 noncritical flags (or exclude those with a critical flag set or five or more noncritical flags) \par }\pard \widctlpar\adjustright {\fs22 \par }\pard\plain \s1\keepn\widctlpar\outlinelevel0\adjustright \b\fs22\cgrid {Results \par }\pard\plain \widctlpar\adjustright \fs20\cgrid {\fs22 \par A total of 134 of the 2363 records were dropped because they are managed sites. A total of 353 appeared to be outliers. We are in the process of reviewing this list. Virtually all the sites identified at the EMDI workshop as outliers were included in this set. If the EMDI group as a whole accepted these as outliers and we recalculate site and biome averages, we would have a total of 1689 records for 933 sites. \par \par }\pard\plain \s1\keepn\widctlpar\outlinelevel0\adjustright \b\fs22\cgrid {Estimating Total NPP \par }\pard\plain \widctlpar\adjustright \fs20\cgrid {\fs22 \par For those sites lacking a value for TNPP, we calculated biome-specific ratios to estimate NPP_EST from ANPP or BNPP. We will recalculate these ratios based on the new biome classification and recalculat e NPP_EST use data with outliers excluded. Based on our analysis, we used a ratio of 0.5 BNPP:TNPP for grasslands, deserts, and tundra. We used a ratio of 0.22 BNPP:TNPP for forests. The RATIO was saved as a variable in the dataset. \par \par }\pard\plain \s1\keepn\widctlpar\outlinelevel0\adjustright \b\fs22\cgrid {Plots and Charts \par }\pard\plain \widctlpar\adjustright \fs20\cgrid {\fs22 \par W e plotted NPP observed against the modeled NPP ensemble, by AET ensemble, and by latitude with the plot symbols indicating the magnitude of the flags. In general, the points that had high flag values appeared to be on the fringe of the cluster of points. As included below, the difference between the old NPP and revised included increases and decreases, generally getting bigger with higher NPP levels. There was a strong relationship of the recalculated model ensemble to be higher at low NPP and lower tha n observed NPP at higher NPP. Bar charts are presented for the new biome groups. \par \par <<>> \par \par \par }{\fs22\cgrid0 The SAS variable names and labels for the flags are: \par \par }\trowd \trgaph108\trleft-108\trbrdrt\brdrs\brdrw10 \trbrdrl\brdrs\brdrw10 \trbrdrb\brdrs\brdrw10 \trbrdrr\brdrs\brdrw10 \trbrdrh\brdrs\brdrw10 \trbrdrv\brdrs\brdrw10 \clvertalt\clbrdrt\brdrs\brdrw10 \clbrdrl\brdrs\brdrw10 \clbrdrb\brdrs\brdrw10 \clbrdrr \brdrs\brdrw10 \cltxlrtb \cellx1620\clvertalt\clbrdrt\brdrs\brdrw10 \clbrdrl\brdrs\brdrw10 \clbrdrb\brdrs\brdrw10 \clbrdrr\brdrs\brdrw10 \cltxlrtb \cellx2430\clvertalt\clbrdrt\brdrs\brdrw10 \clbrdrl\brdrs\brdrw10 \clbrdrb\brdrs\brdrw10 \clbrdrr \brdrs\brdrw10 \cltxlrtb \cellx8640\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 Flag name\cell Flag\cell Flag label\cell }\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 \row }\trowd \trgaph108\trleft-108\trbrdrt\brdrs\brdrw10 \trbrdrl\brdrs\brdrw10 \trbrdrb\brdrs\brdrw10 \trbrdrr\brdrs\brdrw10 \trbrdrh\brdrs\brdrw10 \trbrdrv\brdrs\brdrw10 \clvertalt\clbrdrt\brdrs\brdrw10 \clbrdrl\brdrs\brdrw10 \clbrdrb\brdrs\brdrw10 \clbrdrr\brdrs\brdrw10 \cltxlrtb \cellx1620\clvertalt\clbrdrt\brdrs\brdrw10 \clbrdrl \brdrs\brdrw10 \clbrdrb\brdrs\brdrw10 \clbrdrr\brdrs\brdrw10 \cltxlrtb \cellx2430\clvertalt\clbrdrt\brdrs\brdrw10 \clbrdrl\brdrs\brdrw10 \clbrdrb\brdrs\brdrw10 \clbrdrr\brdrs\brdrw10 \cltxlrtb \cellx8640\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 MIXED_F\cell 100\cell Mixed biome, measurement biome not = to biomenew\cell }\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 \row }\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 BIOME_F\cell 100\cell crops, pasture, plantations, wetlands**\cell }\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 \row }\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 ELEV_MXF\cell 10\cell elev >2500, elev < -100*\cell }\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 \row }\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 PREC_F\cell 10\cell (prec \endash prec_ann) / ave(prec) > 1*\cell }\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 \row }\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 TAVE_F\cell 10\cell (tave \endash temp_ann) / ave(tave) > 1*\cell }\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 \row }\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 MOD_F\cell 10\cell (npp_est - modcb_av) / ave(npp) > 1*\cell }\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 \row }\pard \widctlpar\intbl\adjustright { \fs22\cgrid0 EMDI_F\cell 10\cell Identified at EMDI as an outlier by visual inspection\cell }\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 \row }\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 ANPP_P5F\cell 1\cell anpp outsite .05-.95 percentile by biome2\cell }\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 \row }\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 BNPP_P5F\cell 1\cell bnpp outsite .05-.95 percentile by biome2\cell }\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 \row }\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 TNPP_P5F\cell 1\cell tnpp outsite .05-.95 percentile by biome2\cell }\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 \row }\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 NPP_P5F \cell 1\cell npp outsite .05-.95 percentile by biome2\cell }\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 \row }\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 ANPPBADF\cell 1\cell anpp >2000gc/m2, anpp>95%tnpp\cell }\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 \row }\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 BNPPBADF\cell 1\cell bnpp >2000gc/m2, bnpp>95%tnpp\cell }\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 \row }\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 TNPPBADF\cell 1\cell tnpp >3000gc/m2, anppbadf or bnppbadf\cell }\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 \row }\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 AET_RGF\cell 1\cell NPP outside .95 CI of NPP=a+b*AET regression\cell }\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 \row }\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 PREC_RGF\cell 1\cell NPP outside .95 CI of NPP=a+b*PREC regression\cell }\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 \row }\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 TAVE_RGF\cell 1\cell NPP outside .95 CI of NPP=a+b*TAVE regression\cell }\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 \row }\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 ELEV_F\cell 1\cell (elev_giv - elev_dem) / ave(elev) > 1\cell }\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 \row }\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 PRECBADF\cell 1\cell prec outsite .01-.99 percentile by biome2 (or >1000mm in desert)\cell }\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 \row }\trowd \trgaph108\trleft-108\trbrdrt\brdrs\brdrw10 \trbrdrl\brdrs\brdrw10 \trbrdrb\brdrs\brdrw10 \trbrdrr\brdrs\brdrw10 \trbrdrh \brdrs\brdrw10 \trbrdrv\brdrs\brdrw10 \clvertalt\clbrdrt\brdrs\brdrw10 \clbrdrl\brdrs\brdrw10 \clbrdrb\brdrs\brdrw10 \clbrdrr\brdrs\brdrw10 \cltxlrtb \cellx1620\clvertalt\clbrdrt\brdrs\brdrw10 \clbrdrl\brdrs\brdrw10 \clbrdrb\brdrs\brdrw10 \clbrdrr \brdrs\brdrw10 \cltxlrtb \cellx2430\clvertalt\clbrdrt\brdrs\brdrw10 \clbrdrl\brdrs\brdrw10 \clbrdrb\brdrs\brdrw10 \clbrdrr\brdrs\brdrw10 \cltxlrtb \cellx8640\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 TAVEBADF\cell 1\cell temp outsite .01-.99 percentile by biome2 (>6 in boreal forests)\cell }\pard \widctlpar\intbl\adjustright {\fs22\cgrid0 \row }\pard \widctlpar\adjustright { \par }}