Emulating ecological memory with recurrent neural networks

Basil Kraft*, Simon Besnard, Sujan Koirala

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingChapterAcademicpeer-review

1 Citation (Scopus)


Ecosystem processes are driven both by contemporary and antecedent environmental and land surface conditions through ecological memory effects. This chapter provides an insight into the relevance of memory effects in the Earth system and presents an experimental case study to use an Recurrent Neural Network (RNN) model to emulate a physical model. In addition to introducing an experimental design suitable for such purposes, we demonstrate that an RNN is largely capable of learning the memory effects encoded in a physical model. A non-temporal fully connected model cannot reproduce such memory effects, especially during anomalous conditions (e.g. climate extremes).

Original languageEnglish
Title of host publicationDeep Learning for the Earth Sciences
Subtitle of host publicationA Comprehensive Approach to Remote Sensing, Climate Science and Geosciences
EditorsGustau Camps-Valls, Devis Tuia, Xiao Xiang Zhu, Markus Reichstein
Number of pages13
ISBN (Electronic)9781119646181
ISBN (Print)9781119646143
Publication statusPublished - 20 Aug 2021


  • Deep learning method
  • Earth system
  • Ecological memory effects
  • Evapotranspiration simulations
  • Long short-term memory
  • Recurrent neural network


Dive into the research topics of 'Emulating ecological memory with recurrent neural networks'. Together they form a unique fingerprint.

Cite this