Linje peatland Holocene GDD5 reconstruction
Description
Data for manuscript "First multi-proxy Holocene growing season thermal conditions reconstruction from Central Europe" Poolma et al. 2026 This study aimed to reconstruct Holocene growing season temperatures in Central Europe through a site-specific multi-proxy approach. We established a modern calibration dataset using annually collected Betula nana leaves from the Linje peatland and applied a microphenological inference model to abundant subfossil leaf remains recovered from a 12m long peat core from the same site. In parallel, pollen-based reconstructions derived from the same samples provide an independent estimate of past thermal conditions.
Files
Steps to reproduce
For more detailed info please check the "Methods" section of the original publication and the references within Undulation index of subfossil leaves: Subfossil Betula nana leaves were recovered from the Linje peat core during macrofossil analysis. Samples were collected at 2.5–10 cm intervals throughout the core, spanning depths of 2.5–1187.5 cm. Extracted subfossil B. nana fragments were stored in distilled water at 4 °C until imaging under fluorescent light. Subsequently, the leaves were photographed using a Leica DM6000 B microscope equipped with a Leica DFC310 FX digital colour camera at ×400 magnification. Digital images of subfossil leaves were analysed in ImageJ using a free-hand drawing tool and an XPEN graphic tablet (Model CT430) to measure epidermal pavement cells. For each leaf fragment, at least 13, and ideally 30, pavement cells located in alveolar areas were analysed, while avoiding veins, resin glands, and leaf margins. Cell area (CA) and cell circumference (CC) were measured to calculate the undulation index (UI; dimensionless). UI of different leaf fragments is provided in the table. Pollen analysis: For pollen analysis, 54 peat samples corresponding to those analysed for B. nana leaves were selected from depths between 2.5 and 1147.5 cm. Samples were prepared following standard palynological procedures, including pretreatment with hot 10% potassium hydroxide to remove humic substances, followed by acetolysis. Pollen and spores were counted under a light microscope until a minimum total pollen sum (TPS) of 500 grains per sample was reached. GDD5 reconstruction: Subfossil B. nana leaves – The mean UI value of each sample was used to reconstruct GDD5 using the linear regression model published in the original study. To minimise the influence of analytical variability, the data were also normalised using z-score transformation in RStudio. Pollen – Pollen-based GDD5 reconstructions were performed using the modern pollen analogue dataset of Seppä and Birks (2001). Climate data, including minimum and maximum temperatures (°C), were obtained from the ERA5 dataset with hourly temporal resolution and 0.25° × 0.25° spatial resolution (Hersbach et al., 2023), downloaded from the Copernicus Climate Data Store. Reconstructions were conducted in the C2 software package (Juggins, 2007) using weighted averaging–partial least squares (WA-PLS) regression and calibration (Ter Braak and Juggins, 1993), combined with bootstrapping (9999 permutations) (Birks et al., 1990; Birks, 1998). The optimal transfer function was selected based on the lowest cross-validated root mean squared error of prediction (RMSEP). The data was also normalised using z-score transformation in RStudio.
Institutions
- Tallinn University of TechnologyHarjumaa, Tallinn
- Adam Mickiewicz University in PoznańGreater Poland, Poznan