Chironomids, temperature and numerical models: a reply to Seppälä

Abstract

Holocene comment and reply 615 provide very strong material for statistical analysis. These two samples represent about one third (6.1-9.0°C) of the full temperature range in the study (from 6.1 to 15.4°C).One odd parameter (DBT = distance beyond tree-line in km) is found in Table 1 (Olander et al., 1999). Lakes 24 and 25 should be 81 and 84 km beyond the tree-line, but there are no large tundra areas in Finland which could be so far from the forest limit. Olander et al. (1999) have also measured the pH values of the waters. We must remember that acidity varies in natural waters following the water temperature, and depends on the time of snowmelt, for example. The full range of annual pH variation should be known before we can determine whether it has any effect on the organisms and how this knowledge can be applied in palaeoclimatic reconstructions. A momentary pH measurement provides no indication of the year-round variations of lake acidity. Furthermore the organisms live in the water all year, but which conditions are the critical determinants? Ice coverThe lakes on the high fells remain ice-covered often long into spring, but this does not necessarily mean that the air temperatures there are lower in winter. The ice cover on these lakes only becomes thicker because wind drifts insulating snow off the ice surface. In spring, more solar energy is needed to melt the thick lake ice. Ice thickness in this environment can be more than 1 m. This surely has an effect on the light requirements and living conditions of the aquatic organisms. Changes in lake-ice thickness need not reflect changes in air temperatures but may be controlled by snow depth and/or wind conditions. Some of the lakes do not become ice-free before the beginning of July. Therefore one may ask why should the occurrences of chironomid taxa be correlated with July mean temperatures if the most suitable living period might be August, even though the July air temperature is the highest. When modelling the living conditions of cladocera or chironomids, these types of factors should be considered. 'The primary climate effect on the aquatic biota of high-altitude lakes may be mediated by the timing of the ice cover' as Livingstone et al. (1999) warned.The goal of Olander et al.'s (1999) study was really ambitious. 'Particular attention was paid to assessing the power of lakewater temperature in explaining the variance in the chironomid data, as the primary aim is to develop a chironomid-based calibration function for palaeotemperature reconstructions' (p. 284). It contains much new information on the chironomid taxa in subarctic waters in Lapland, but unfortunately the data used fail to provide a satisfactory basis for the statistical analyses. The conclusions drawn from these statistical evaluations are built on very weak foundations and are thus unreliable and probably misleading. (2001) concerning the suitability of the surface-water temperature measurements and mean July air temperature estimates used by Olander et al. (1999) to derive quantita...

A chironomid-based reconstruction of summer temperatures in NW Iceland since AD 1650

“…The use of transfer functions for reconstructing single environmental variables from biotic proxies (i.e. climate in this instance) has long been a matter of debate (see Korhola et al, 2001 and references therein). It is not enough to simply create transfer functions which have good predictive powers, the real proof is in model validation, for it is well recognised that species almost always respond to several interacting forces which may cascade down the food-web, leading to modelled species-environment relationships being often indirect (see Sayer et al, 2010 for a full discussion).…”

John Birks: Pioneer in quantitative palaeoecology

“…During the 1990s, many key papers were published describing and analysing modern calibration training sets that did not include fossil data or environmental reconstructions. Some of these studies caused a lively debate about organism-environment relationships in arctic regions (Olander et al, 1999; Korhola et al, 2001). A need for a similar modern calibration model paper for pollen data was recognised and such a paper was consequently published by Seppä et al (2004).…”

“…Some of these studies caused a lively debate about organism-environment relationships in arctic regions (Olander et al, 1999;Korhola et al, 2001). A need for a similar modern calibration model paper for pollen data was recognised and such a paper was consequently published by Seppä et al (2004).…”

Revealing the Emperor's new clothes: niche-based palaeoenvironmental reconstruction in the light of recent ecological theory

“…Usually, the reconstruction is limited to taxa from a particular fossil group (eg, diatoms, testate amoebae) thought to show strong niche partitioning along the Birks (1995) summarizes the five main assumptions that underpin quantitative palaeoenvironmental reconstruction (Table 1). The validity of some of these assumptions has been debated previously (eg, Warner and Hann, 1987;Korhola et al, 2001). In particular, the concepts of 'inertia' and 'threshold' (Smith, 1965;Cole, 1985;Carrión et al, 2001), as well as 'no modern analogue' assemblages (Overpeck et al, 1985;Jackson and Williams, 2004), describe non-linear dynamics and historical contingencies that lead to uncertainty in palaeoenvironmental interpretation and reconstruction.…”