<div class="csl-bib-body">
<div class="csl-entry">Hobeichi, S., Shao, Y., Rampal, N., Bittner, M., & Abramowitz, G. (2024, April 17). <i>Revisiting Tabular Machine Learning and Sequential Models to Advance Climate Downscaling</i> [Conference Presentation]. EGU24 General Assembly, Wien, Austria. https://doi.org/10.5194/egusphere-egu24-7111</div>
</div>
-
dc.identifier.uri
http://hdl.handle.net/20.500.12708/208632
-
dc.description.abstract
Recent advancements in the empirical downscaling of climate fields using Machine Learning have predominantly leveraged computer vision approaches. These methods treat a climate field as an image channel, applying image processing techniques to automatically extract features for the downscaling model from its latent space embeddings. In contrast, this work aims to revisit and validate the potential of tabular and sequential models in the context of grid-by-grid downscaling, where each grid cell in a map is individually downscaled and input features for the downscaling model are selected manually by a climate expert. We present downscaling results for precipitation and evapotranspiration using three distinct models: Long Short-Term Memory (LSTM), Multi-layer Perceptron (MLP), and a hybrid approach that combines Linear Regression with Random Forest. Our discussion includes the setup and optimization strategies for these models to enhance their ability to capture extremes. The merits of this grid-by-grid approach are highlighted, focusing not only on performance and effectiveness in preserving spatial features but also on its flexibility, spatial transferability, ease of model fine-tuning, and training efficiency.
en
dc.description.sponsorship
Christian Doppler Forschungsgesells
-
dc.language.iso
en
-
dc.subject
Climate Downscaling
en
dc.subject
Sequential Models
en
dc.subject
Tabular Machine Learning
en
dc.title
Revisiting Tabular Machine Learning and Sequential Models to Advance Climate Downscaling
en
dc.type
Presentation
en
dc.type
Vortrag
de
dc.contributor.affiliation
National Institute of Water and Atmospheric Research, New Zealand
-
dc.relation.grantno
123456
-
dc.type.category
Conference Presentation
-
tuw.project.title
CDL Embedded Machine Learning
-
tuw.researchTopic.id
E4
-
tuw.researchTopic.name
Environmental Monitoring and Climate Adaptation
-
tuw.researchTopic.value
100
-
tuw.publication.orgunit
E384-02 - Forschungsbereich Systems on Chip
-
tuw.publisher.doi
10.5194/egusphere-egu24-7111
-
tuw.author.orcid
0000-0001-6825-3854
-
tuw.author.orcid
0000-0002-9938-669X
-
tuw.author.orcid
0000-0001-9801-9348
-
tuw.author.orcid
0009-0004-8022-2232
-
tuw.author.orcid
0000-0002-4205-001X
-
tuw.event.name
EGU24 General Assembly
en
tuw.event.startdate
14-04-2024
-
tuw.event.enddate
19-04-2024
-
tuw.event.online
Hybrid
-
tuw.event.type
Event for scientific audience
-
tuw.event.place
Wien
-
tuw.event.country
AT
-
tuw.event.presenter
Hobeichi, Sanaa
-
wb.sciencebranch
Elektrotechnik, Elektronik, Informationstechnik
-
wb.sciencebranch.oefos
2020
-
wb.sciencebranch.value
100
-
item.grantfulltext
none
-
item.openairetype
conference paper not in proceedings
-
item.openairecristype
http://purl.org/coar/resource_type/c_18cp
-
item.languageiso639-1
en
-
item.fulltext
no Fulltext
-
item.cerifentitytype
Publications
-
crisitem.author.dept
National Institute of Water and Atmospheric Research, New Zealand
