Holt Forest Timber Inventory Notes on the Data Tracking 32,000 individual trees over 35 years is tricky; the individual tree data should be used with caution. For instance there are 125 non-trivial duplicate TREENOs, most of which can be assumed to be real data from a misrecorded tree. Missing Values Some missing values ('NA') are just missing but some patterns are meaningful. Here are some examples: "YEAR","QUAD","SUBQUAD","TREENO","SPEC","FIASP","DBH","COND" 1996,"3F4",2,NA,1,129,NA,6 # Condition=6 is "dead-and-down", so any values might be missing and DBH=NA # is correct -- no dbhs should be taken for dead-and-down. 1988,"3F4",2,151.1,1,129,38,0 1988,"3F4",2,151.2,1,129,34,0 # and subsequently... 1996,"3F4",2,151.1,1,129,59,0 1996,"3F4",2,151.2,1,129,NA,0 # Multi-stem trees (TREENOs with decimals) are a particular bookkeeping # problem. When stems merge we try to keep all TREENOs, give the dbh to the # dominant and NAs for the rest. This pattern is not perfectly consistent but # it's the intention. 1986,"3E1",1,3,1,129,19,0 1988,"3E1",1,3,1,129,NA,NA 1996,"3E1",1,3,1,129,21,0 2007,"3E1",1,3,1,129,21,0 # Missing DBH and COND is for 'logical trees' -- those that were apparently # overlooked in an inventory but must have been there, so have been inserted # without measurements. 1988,"6D2",2,945,1,129,11,0 1996,"6D2",2,945,5,833,20,0 # becomes 1988,"6D2",2,945,5,833,NA,NA 1988,"6D2",2,NA,1,129,11,0 1996,"6D2",2,945,5,833,20,0 # Missing TREENO can result from something like the above, where an inexplicable # error is combined with a positive correction.