SCIENTIFIC METHODOLOGY OF RECONSTRUCTING CLIMATE

            The means exist to reconstruct the climate beyond historic documentation.  Through paleoclimatology, seasonal precipitation, and average temperature can be extrapolated over the span of the proxy indicator.  Through the Principle of Uniformitarianism, and the idea that biological responses in the past are the same as the present, it is possible to use scientific methods to accurately reconstruct climate change (Bradley, 1999) .  The only limitations of paleoclimatology are the availability and integrity of samples. 

The process of reconstructing the climatological conditions of the first few centuries AD in Germania allows the building a model of modern agricultural responses, determining the social structure of the inhabitants, interpreting the data compiled from various “proxies” (indicators) of climate, and applying the paleoclimatic interpretation to antiquity.  The botanical responses are assumed to be consistent in both antiquity and modern times, presuming that the conditions observed are similar in both eras (Fritts, 1976). According to Parry (1981), “Climate Change and the Agricultural Frontier”, the first stage in the modeling the process of paleoclimatology (e.g., dendrochronology and palynology) starts with an analogous crop-climate relationship of the present (Parry, 1981) .  The crop-climate conditions are compared to the interaction of variable weather and various farming methods and decisions.  The meteorological data from the modern era is related to modern agricultural productivity and this relationship is used to reconstruct variation in agroclimates.  The responses to these variations are determined and quantified according to the different plant species.  These climate variations are monitored in degree of severity and duration to determine the overall agricultural responses (Parry, 1981) .  These responses are the basis of modeling the agro-climate structures to explain plant growth response to climate.

Fritts (1976) describes the process of dendroclimatological reconstruction through six phases: (1) plant process, (2) operational environment, (3) microclimate, (4) weather, (5) climate state, and (6) macroclimate (Fritts, 1976) .  Upon understanding one phase, the subsequent phases become comprehensible.  The (1) plant process deals with the physiological processes of the plants and the way in which their response to climatic variation is made evident. 

Upon understanding of the plant processes, the (2) operational environment can be examined.  The operational environment examines the plant as a whole, concentrating on the plant and its response to climate variables.  When the operational environment of the plants is understood, accurate reconstruction of regional climate is achievable.  This is due to the Principle of Uniformity in the order of nature.  The responses of vegetation to (3) climate conditions in the present are assumed the same in antiquity.  Potentially, the variation in plant growth, or any other (4) environmental variable that can be interpreted in terms of instrumental data, are proxies of climate.

When examining ancient history, the last two of Fritts’ phases are the most important: the (5) climate state and (6) macroclimate.  The climate state uses the information gathered about weather conditions to assess the climatic conditions of the entire region over a period of months or years, as in this study of Germania.  The study of macroclimate examines the climate state over many years and generalizes the information (i.e., the macroclimate of Germania in AD 200 to AD 400 was a period of drought and poor agricultural conditions although there were certainly years, and perhaps even decades of favorable conditions interspersed in these centuries.)

Retrospection is the key to understanding climate indicators in biological data sets, such as flora and fauna, as well as meteorological data.  Biological data can be used as climate indicators due to the consistency of their response under the similar climate conditions.  However, biological data gathered in very recent times is difficult to relate to climate variation due to Global Warming and advances in modern agriculture.

The overlapping of several climate indicators helps ameliorate the complications of variable interpretations rendered from individual climate indicators.  For example, a single dendrochronological sample can command several valid interpretations, all of which are attached to the growth of a particular tree species in a region, but the use of several dendrochronological studies and/or other climate indicators may increase accuracy of interpretation.

Biological indicators, such as tree rings and pollen, are the foundations of paleoclimatological studies, because they examine the plant process and operational environment phases, which are the building blocks of all interpretation.  Biological proxies can be highly predictable and therefore subject to scientifically based reconstruction.  It is also useful to incorporate stable isotopic data and sedimentary indicators (e.g., riverbed sediments or ice core samples) to enhance biologically based interpretations of past climate.  It is dangerous to rely upon a single proxy, even dendrochronology, to interpret climate change.  It would be as if the historian relied upon a single document to reconstruct the history of a people.

            In antiquity, biological indicators are important for identifying the timing and severity of agricultural anomalies.  Dendrochronological reconstruction, ice cores, and varves are among the few indicators that can potentially reveal year-to-year changes in agricultural yields and growth potential.  There are several indicators, such as palynology and lake and ocean sediments, that can be used to reconstruct vegetation, temperature, and rainfall, but they are often limited to general seasonal changes and lack the annual precision of dating afforded by tree rings.  Still other types of data: speleothems, variation of oxygen isotopes in coral, radiocarbon records of sunspot activity, peat cellulose studies, and other types of isotopic data that can reveal climate changes over decades, centuries, and millennia (Linacre, 1992) .  These different types of indicators, revealing data chronologically accurate to a single year or to an approximate date prove useful in reconstructing ancient climates.