How Well Understood Are the Processes that Create Dendroclimatic Records? A Mechanistic Model of the Climatic Control on Conifer Tree-Ring Growth Dynamics - Dendroclimatology - page 38

models are increasingly valuable (McCarroll and Loader 2004 ) , the most widely
used tree-ring characteristics for climate reconstructions have been tree-ring width
and maximum latewood density. Such reconstructions have relied almost entirely on
statistical covariation of the proxy observations and climate data during the modern
period.
But Harold C. Fritts saw the vital importance of concurrent research into the
anatomical pattern of tree-ring formation in response to changing environment. He
was one of the first to make a careful study of tree-ring formation in ponderosa
pine on the anatomical level, showing how the current and previous weather con-
ditions affect the seasonal growth and cell dimensions of tree rings (Fritts 1966 ,
1976 ) . Studies like this have provided broad knowledge about which processes in
the growing tree control cell production and which determine the characteristics of
those cells which are produced. Briefly, tree-ring formation occurs in the active veg-
etation season, and the processes involved in tree-ring formation integrate the prior
and current seasonal conditions. Growth starts with cambial zone activation (acti-
vation of the fast divisions of cells). Newly produced cells progress to enlargement,
and finally to cell wall thickening. At this stage the cell loses the protoplast and
transforms to the 'dead' element of the water conductive system of the tree.
The focus of much anatomical research since Fritts' early work has been on cam-
bial activity (the workings of the 'cambial machine'). Detailed observations and
analyses of basic processes of wood formation allowed Wilson and Howard ( 1968 ) ,
Howard and Wilson ( 1972 ) , and Wilson ( 1973 ) to offer a 'mimic' model for the
quantitative description of tree-ring formation. While this model reproduced the
structures formed in the tree stem during the growing season, it required an enor-
mous number of variables and seasonal observational data that had to be specifically
defined for any particular tree ring. Furthermore, the model did not include exter-
nal environmental parameters. Despite the lack of explicit modeling of the external
environmental influence on the processes of tree-ring growth, or perhaps because
of it, the nature of the environmental control of wood formation has often simply
been assumed. The implicit hypothesis has been that environmental factors influence
tree-ring formation indirectly, through hormone and substrate balance produced by
the apical meristems of shoots, and probably roots (Larson 1964 ; Denne and Dodd
1981 ) .
3.1.1 The Substrate Source-Sink Hypothesis
In his pioneering work on the biological basis for dendroclimatology, Fritts ( 1976 )
included extensive descriptions of the role of both metabolic and cambial processes
as mediators of the environmental control on annual tree growth and ring formation.
Metabolic approaches often rely on a source-sink understanding of the influence
of climate on tree-ring width, which assumes that the environment indirectly deter-
mines annual xylem increment as a function of the amount of carbon available for
secondary growth. Because secondary basal growth is interpreted to be a low priority
for the apportionment of the available whole plant resources (Savidge 2000 a ), wider