B

T~T1

FIGURE 10.8 Cross dating of tree rings. Comparison of tree-ring widths makes it possible to identify false rings or where rings are locally absent. For example, in (A) strict counting shows a clear lack of synchrony in the patterns. In the lower specimen of (A), rings 9 and 16 can be seen as very narrow, and they do not appear at all in the upper specimen. Also, rings 21 (lower) and 20 (upper) show intra-annual growth bands. In (B), the positions of inferred absence are designated by dots (upper section), the intra-annual band in ring 20 is recognized, and the patterns in all ring widths are synchronously matched (Fritts, 1976).

horizons (e.g., those associated with a volcanic event of known age), dating of such proxies is always more uncertain than in dendroclimatic studies (Stahle, 1996).

Once the samples have been cross-dated and a reliable chronology has been established there are three important steps to produce a dendroclimatic reconstruction:

1. standardization of the tree ring parameters to produce a site chronology;

2. calibration of the site chronology with instrumentally recorded climatic data, and production of a climatic reconstruction based on the calibration equations; and

3. verification of the reconstruction with data from an independent period not used in the initial calibration.

In the next three sections, each of these steps is discussed in some detail.

10.2.3 Standardization of Ring-Width Data

Once the chronology for each core has been established, individual ring widths are measured and plotted to establish the general form of the data (Fig. 10.9). It is common for time series of ring widths to contain a low frequency component resulting entirely from the tree growth itself, with wider rings generally produced during the early life of the tree. In order that ring-width variations from different cores can be

Ring Measurements and Standardized Indices

Comparison

0 0

Post a comment