Northern Russian chironomid-based modern summer temperature data set and inference models

West and East Siberian data sets and 55 new sites were merged based on the high taxonomic similarity, and the strong relationship between mean July air temperature and the distribution of chironomid taxa in both data sets compared with other environmental parameters. Multivariate statistical analysi...

Full description

Bibliographic Details
Published in:Global and Planetary Change
Main Authors: Nazarova, Larisa, Self, Angela E., Brooks, Stephen J., van Hardenbroek, Maarten, Herzschuh, Ulrike, Diekmann, Bernhard
Format: Article in Journal/Newspaper
Language:English
Published: 2014
Subjects:
Online Access:https://eprints.soton.ac.uk/378300/
https://eprints.soton.ac.uk/378300/1/Nazarova_20et_20al_20in_20press_20Russian_20combined_20TF.pdf
Description
Summary:West and East Siberian data sets and 55 new sites were merged based on the high taxonomic similarity, and the strong relationship between mean July air temperature and the distribution of chironomid taxa in both data sets compared with other environmental parameters. Multivariate statistical analysis of chironomid and environmental data from the combined data set consisting of 268 lakes, located in northern Russia, suggests that mean July air temperature explains the greatest amount of variance in chironomid distribution compared with other measured variables (latitude, longitude, altitude, water depth, lake surface area, pH, conductivity, mean January air temperature, mean July air temperature, and continentality). We established two robust inference models to reconstruct mean summer air temperatures from subfossil chironomids based on ecological and geographical approaches. The North Russian 2-component WA-PLS model (RMSEPJack = 1.35 °C, rJack2 = 0.87) can be recommended for application in palaeoclimatic studies in northern Russia. Based on distinctive chironomid fauna and climatic regimes of Kamchatka the Far East 2-component WAPLS model (RMSEPJack = 1.3 °C, rJack2 = 0.81) has potentially better applicability in Kamchatka.