English
 
User Manual Privacy Policy Disclaimer Contact us
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT
  Identifying dynamic memory effects on vegetation state using recurrent neural networks

Kraft, B., Jung, M., Körner, M., Requena Mesa, C., Cortés, J., & Reichstein, M. (in press). Identifying dynamic memory effects on vegetation state using recurrent neural networks. Frontiers in Big Data.

Item is

Basic

show hide
Item Permalink: http://hdl.handle.net/21.11116/0000-0004-8A54-2 Version Permalink: http://hdl.handle.net/21.11116/0000-0004-9818-6
Genre: Journal Article

Files

show Files

Locators

show

Creators

show
hide
 Creators:
Kraft, Basil1, 2, Author              
Jung, Martin1, Author              
Körner, Marco, Author
Requena Mesa, Christian2, 3, Author              
Cortés, José2, Author              
Reichstein, Markus4, Author              
Affiliations:
1Global Diagnostic Modelling, Dr. Martin Jung, Department Biogeochemical Integration, Dr. M. Reichstein, Max Planck Institute for Biogeochemistry, Max Planck Society, ou_1938311              
2IMPRS International Max Planck Research School for Global Biogeochemical Cycles, Max Planck Institute for Biogeochemistry, Max Planck Society, Hans-Knöll-Str. 10, 07745 Jena, DE, ou_1497757              
3Empirical Inference of the Earth System, Dr. Miguel D. Mahecha, Department Biogeochemical Integration, Dr. M. Reichstein, Max Planck Institute for Biogeochemistry, Max Planck Society, ou_1938312              
4Department Biogeochemical Integration, Dr. M. Reichstein, Max Planck Institute for Biogeochemistry, Max Planck Society, ou_1688139              

Content

show
hide
Free keywords: -
 Abstract: Vegetation state is largely driven by climate and the complexity of involved processes leads to non-linear interactions over multiple time-scales. Recently, the role of temporally lagged dependencies, so-called memory effects, has been emphasized and studied using data-driven methods, relying on a vast amount of Earth observation and climate data. However, the employed models are often not able to represent the highly non-linear processes and do not represent time explicitly. Thus, data-driven study of vegetation dynamics demands new approaches that are able to model complex sequences. The success of Recurrent Neural Networks (RNNs) in other disciplines dealing with sequential data, such as Natural Language Processing, suggests adoption of this method for Earth system sciences. There are only a few recent studies that used the mentioned models to predict Earth system variables, and none that explore the potential of RNNs as a tool to better understand the temporal scales on which environmental conditions affect vegetation state. Here, we used a Long Short-Term Memory (LSTM) architecture to fit a global model for Normalized Difference Vegetation Index (NDVI), a proxy for vegetation state, by using climate time-series and static variables representing soil properties and land cover as predictor variables. Furthermore, a set of permutation experiments was performed with the objective to identify memory effects and to better understand the scales on which they act under different environmental conditions. This was done by comparing models that have limited access to temporal context, which was achieved through sequence permutation during model training. We performed a cross-validation with spatio-temporal blocking to deal with the auto-correlation present in the data and to increase the generalizability of the findings. With a full temporal model, global NDVI was predicted with R2 of 0.943 and RMSE of 0.056. The temporal model explained 14% more variance than the non-memory model on global level. The strongest differences were found in arid and semiarid regions, where the improvement was up to 25%. Our results show that memory effects matter on global scale, with the strongest effects occurring in sub-tropical and transitional water-driven biomes.

Details

show
hide
Language(s):
 Dates: 2019-08-22
 Publication Status: Accepted / In Press
 Pages: -
 Publishing info: -
 Table of Contents: -
 Rev. Method: -
 Identifiers: Other: BGC3123
 Degree: -

Event

show

Legal Case

show

Project information

show

Source 1

show
hide
Title: Frontiers in Big Data
  Abbreviation : Front. Big Data
Source Genre: Journal
 Creator(s):
Affiliations:
Publ. Info: Lausanne : Frontiers Media
Pages: - Volume / Issue: - Sequence Number: - Start / End Page: - Identifier: ISSN: 2624-909X
CoNE: https://pure.mpg.de/cone/journals/resource/2624-909X