Accurate vegetation models can produce further insights into the complex inter-
action between vegetation activity and ecosystem processes. Previous research
has established that long-term trends and short-term variability of temperature
and precipitation affect vegetation activity. Motivated by the recent success of
Transformer-based Deep Learning models for medium-range weather forecast-
ing, we adapt the publicly available pre-trained FourCastNet to model vegetation
activity while accounting for the short-term dynamics of climate variability. We
investigate how the learned global representation of the atmosphere’s state can
be transferred to model the normalized difference vegetation index (NDVI). Our
model globally estimates vegetation activity at a resolution of 0.25◦ while relying
only on meteorological data. We demonstrate that leveraging pre-trained weather
models improves the NDVI estimates compared to learning an NDVI model from
scratch. Additionally, we compare our results to other recent data-driven NDVI
modeling approaches from machine learning and ecology literature. We further
provide experimental evidence on how much data and training time is necessary
to turn FourCastNet into an effective vegetation model. Code and models will be
made available upon publication.
Описание
[2403.18438v1] Global Vegetation Modeling with Pre-Trained Weather Transformers
%0 Generic
%1 janetzky2024global
%A Janetzky, Pascal
%A Gallusser, Florian
%A Hentschel, Simon
%A Hotho, Andreas
%A Krause, Anna
%D 2024
%K author:GALLUSSER author:HOTHO author:JANETZKY author:KRAUSE ecoapplication ecomodelling myown
%T Global Vegetation Modeling with Pre-Trained Weather Transformers
%U https://arxiv.org/abs/2403.18438v1
%X Accurate vegetation models can produce further insights into the complex inter-
action between vegetation activity and ecosystem processes. Previous research
has established that long-term trends and short-term variability of temperature
and precipitation affect vegetation activity. Motivated by the recent success of
Transformer-based Deep Learning models for medium-range weather forecast-
ing, we adapt the publicly available pre-trained FourCastNet to model vegetation
activity while accounting for the short-term dynamics of climate variability. We
investigate how the learned global representation of the atmosphere’s state can
be transferred to model the normalized difference vegetation index (NDVI). Our
model globally estimates vegetation activity at a resolution of 0.25◦ while relying
only on meteorological data. We demonstrate that leveraging pre-trained weather
models improves the NDVI estimates compared to learning an NDVI model from
scratch. Additionally, we compare our results to other recent data-driven NDVI
modeling approaches from machine learning and ecology literature. We further
provide experimental evidence on how much data and training time is necessary
to turn FourCastNet into an effective vegetation model. Code and models will be
made available upon publication.
@misc{janetzky2024global,
abstract = {Accurate vegetation models can produce further insights into the complex inter-
action between vegetation activity and ecosystem processes. Previous research
has established that long-term trends and short-term variability of temperature
and precipitation affect vegetation activity. Motivated by the recent success of
Transformer-based Deep Learning models for medium-range weather forecast-
ing, we adapt the publicly available pre-trained FourCastNet to model vegetation
activity while accounting for the short-term dynamics of climate variability. We
investigate how the learned global representation of the atmosphere’s state can
be transferred to model the normalized difference vegetation index (NDVI). Our
model globally estimates vegetation activity at a resolution of 0.25◦ while relying
only on meteorological data. We demonstrate that leveraging pre-trained weather
models improves the NDVI estimates compared to learning an NDVI model from
scratch. Additionally, we compare our results to other recent data-driven NDVI
modeling approaches from machine learning and ecology literature. We further
provide experimental evidence on how much data and training time is necessary
to turn FourCastNet into an effective vegetation model. Code and models will be
made available upon publication.},
added-at = {2024-04-12T08:39:17.000+0200},
archiveprefix = {arXiv},
author = {Janetzky, Pascal and Gallusser, Florian and Hentschel, Simon and Hotho, Andreas and Krause, Anna},
biburl = {https://www.bibsonomy.org/bibtex/26bb1e4a493e91322164a06316a041946/jascal_panetzky},
description = {[2403.18438v1] Global Vegetation Modeling with Pre-Trained Weather Transformers},
eprint = {2403.18438},
interhash = {8b33799de36fa4d973eeacf5e17cde8d},
intrahash = {6bb1e4a493e91322164a06316a041946},
keywords = {author:GALLUSSER author:HOTHO author:JANETZKY author:KRAUSE ecoapplication ecomodelling myown},
primaryclass = {cs.LG},
timestamp = {2024-04-12T08:39:17.000+0200},
title = {Global Vegetation Modeling with Pre-Trained Weather Transformers},
url = {https://arxiv.org/abs/2403.18438v1},
year = 2024
}