METEOROLOGICAL DATA AND TREE RINGS

Paleoclimatological data from Ancient World times are significantly scarcer than from Medieval and modern times (Pavese, 1992) .  This makes reconstructing the climate of ancient times more difficult than the recent past.  However, data are becoming increasingly available as more researchers publish their findings and as access to previously unobtainable data improves.

As with any reconstructions, it is necessary to reason backwards, or “retrodict”, to understand climate indicators and climate change (Wigley et al., 1981, Lamb, 1982) .  “Climate” is the total experience of weather at any place over some specific period (Lamb, 1977) .  Proxy indicators such as flora and fauna as well as meteorological data can only be understood by applying modern experience to them (Wigley et al., 1981) .

If it is necessary to understand the meteorological instrumental record, it is necessary to understand how the data were recorded.  Just because the meteorological data have been recorded does not mean that the data were accurate or the instrumentation was the same (Linacre, 1992) .  It is necessary to understand the characteristics of the older meteorological instruments, the nature of exposure of the instruments, the time that the data were recorded, know the scales used to measure the changes, and know the location of the samples (Wigley et al., 1981) .  Meteorological data sources have varied over time, and it is important to know if the earlier data (late AD 1700-1900) has been properly recorded/ converted for use with the more recent data.  Some of the earliest meteorological stations that will be used in this paper are Berlin (1719 to present with a break for the World Wars), Basel (1755 to present), Geneva (1768 to present), Vienna (1775 to present), Höhenpeißenberg (1781 to present with a break in World War II), and Great Saint Bernard in the Alps (1818 to present) (Pfister and Lauterburg, 1992, Schuepp and Schirmer, 1977, Chernavskaya, 1996, World Meteorological Organization. and National Climatic Data Center (U.S.), 1998) .

Arguably, dendroclimatology is the best-known paleoclimatic proxy.  With the introduction of dendrochronology to Europe by Bruno Huber (1940’s), the amount of data has increased exponentially over the years (Brongers, 1973) .  Granted, Huber had to modify the methodology of A. E. Douglass of the USA to methodologies that were more appropriate for the shorter sample lengths of European trees but through the methodology of standardization and gleichlaufigkeitswert (sometimes just called gleichläufigkeit) (GLK) was able to confidently construct chronologies in central Europe (Huber, 1943) .  The development of GLK coefficient method was of utmost importance in the formation of the central European chronologies, and with the use of computer technology, this time consuming process has been made more efficient.

It is well known that trees grow at variable, but predictable, rates over time (Glock and Pearson, 1937) .  Trees tend to produce smaller rings as they age, so through “standardization”, this can be equalized so that the growth trends can be examined in many trees at the same scale from the same area (Fritts, 1976) .  As discussed earlier, there can be difficulties in dendrochronological interpretation with low data density or sample depth, which can make it difficult to separate various components of climate (temperature and precipitation) (Guiot, 1992) , but how well do tree-ring data of one area in Germania compare to other areas? 

Making a direct comparison between modern meteorological data and tree rings, it would appear that in modern times climate is stabilizing or the climate signal recorded in the tree rings is becoming less sensitive.  This appearance may come from modern use of hybrids to counter harsh environments and these hybrids can accidentally be brought into the tree-ring record, giving it a false or non-climatically driven stability (Fritts, 1976) .  It is important to remain conservative when evaluating paleoclimatological data, even recorded instrumental data, for assurance that the data are accurate and pertinent.