Testing evidence for savanna corridors in South-East Asia since the Last Glacial Maximum
Analysis of palaeoecological records across South-East Asia to determine the evidence for regime shifts between open savannas and dense tropical forests occurred since the Last Glacial Maximum
Code contributors: Rebecca Hamilton, Corey J. A. Bradshaw, Frédérik Saltré, Jesse Wolfhagen
Accompanies paper (in review):
Hamilton, R, N Amano, CJA Bradshaw, F Saltré, R Patalano, D Penny, J Stevenson, J Wolfhagen, P Roberts. 2023. Seasonal forests, not long-term savanna corridors, dominated in South-East Asia during the Last Glacial Maximum.
Abstract
We did new qualitative and statistical analysis of 59 palaeoecological records across South-East Asia to show that, instead of swings between open savanna and dense, ever-wet rain forest ecosystems, the climatic changes of the Last Glacial Period and particularly the Last Glacial Maximum, involved more fluid transitions between lowland rainforest, seasonally dry ‘savanna’ forest, and tropical montane forest. This ‘hybrid’ forest-savanna biome provides a route beyond currently accepted notions of binary ecologies for the region, and yields new insights into ecological resilience of this region in the past and future.
Scripts
Data
Site-specific palaeoecological proxy series (raw data)
- G6-4.csv
- SH19014_grass.csv
- G5_6_149P2.csv
- NPK2.csv
- hordorli.csv
- 18300.csv
- 18323.csv
- 18302.csv
- CB19.csv
- MD063075.csv
- NS-0725.csv
- 17964.csv
- DDA.csv
- G4_K12P1.csv
- PB-A.csv
- GEOB100693.csv
- PSS.csv
- GeoB100537.csv
- G5_2_056P.csv
- LL2.csv
- RD-3.csv
- KUM3.csv
- G4_K4P3.csv
- BYK2.csv
- TOW9.csv
- BJ8_03_91GGC.csv
- MAT10_2B.csv
- SO189_144KL.csv
- MC1.csv
- mbelen.csv
Compiled, age-resampled, standardised compilation of time series (34 ka to 2 ka), with site coordinates (lon/lat)
- SC_all_SACor.csv
Spatially constrained bivariate correlations to reproduce site-by-site correlation heatmap
- cor_matrix_allmean.csv
Supplementary data
- database descriptors (Excel file)
R packages
Scripts require following R libraries
spatstat
gstat
maps
sp
ape
permute
ggplot2
dplyr
boot
tmvnsim
wCorr
hrbrthemes