Preprints
https://doi.org/10.5194/egusphere-2024-2128
https://doi.org/10.5194/egusphere-2024-2128
09 Aug 2024
 | 09 Aug 2024

A novel explainable deep learning framework for reconstructing South Asian palaeomonsoons

Kieran M. R. Hunt and Sandy P. Harrison

Abstract. We present novel explainable deep learning techniques for reconstructing South Asian palaeomonsoon rainfall over the last 500 years, leveraging long instrumental precipitation records and palaeoenvironmental datasets from South and East Asia to build two types of model: dense neural networks ('timeline models') and convolutional neural networks (CNNs). The timeline models are trained individually on seven regional rainfall datasets and while they capture decadal-scale variability and significant droughts, they underestimate interannual variability. The CNNs, designed to account for spatial relationships in both predictor and target, demonstrate higher skill in reconstructing rainfall patterns and produce robust spatiotemporal reconstructions. The 19th and 20th centuries were characterised by marked inter-annual variability in the monsoon, but earlier periods were characterised by more decadal- to centennial-scale oscillations. Multidecadal droughts occurred in the mid-seventeenth and nineteenth centuries, while much of the eighteenth century (particularly the early part of the century) was characterised by above-average monsoon precipitation. Extreme droughts tend to be concentrated in south and west India and often coincide with recorded famines. By applying explainability techniques, we show that the models make use of both local hydroclimate and synoptic-scale dynamical relationships. Our findings offer insights into the historical variability of the Indian summer monsoon and highlight the potential of deep learning techniques in palaeoclimate reconstruction.

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this preprint. The responsibility to include appropriate place names lies with the authors.
Kieran M. R. Hunt and Sandy P. Harrison

Status: final response (author comments only)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on egusphere-2024-2128', Anonymous Referee #1, 30 Aug 2024
  • RC2: 'Comment on egusphere-2024-2128', Nick Scroxton, 10 Sep 2024
Kieran M. R. Hunt and Sandy P. Harrison

Data sets

Data-driven reconstructions of the Indian palaeomonsoon (1500–1995 CE) Kieran M. R. Hunt and Sandy P. Harrison https://zenodo.org/records/12688184

Kieran M. R. Hunt and Sandy P. Harrison

Viewed

Total article views: 392 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
271 82 39 392 9 5
  • HTML: 271
  • PDF: 82
  • XML: 39
  • Total: 392
  • BibTeX: 9
  • EndNote: 5
Views and downloads (calculated since 09 Aug 2024)
Cumulative views and downloads (calculated since 09 Aug 2024)

Viewed (geographical distribution)

Total article views: 407 (including HTML, PDF, and XML) Thereof 407 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 01 Nov 2024
Download
Short summary
In this study, we train machine learning models on tree rings, speleothems, and instrumental rainfall to estimate seasonal monsoon rainfall over India over the last 500 years. Our models highlight multidecadal droughts in the mid-seventeenth and nineteenth centuries, and we link these to historical famines. Using techniques from explainable AI, we show our models use known relationships between local hydroclimate and the monsoon circulation.