Machine Learning for Applied Weather Prediction

Haupt, Sue Ellen; Cowie, J. M. G.; Linden, Seth; McCandless, Tyler; Kosović, Branko; Alessandrini, Stefano

doi:10.1109/escience.2018.00047

Cited by 40 publications

(33 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The IEEE eScience Conference session on weather and climate science included presentations of state-of-the art research at the interface of weather and climate science and digital technologies. Contributions were selected after a peer review on their scientific merit and innovative nature and published in the conference proceedings (Bari; Behrens et Map and openly available in-situ meteorological observations (Haupt et al, 2018;Garcia-Marti et al, 2018;Bari, 2018;Schultz et al, 2018, and references therein). Also, citizen data like social media posts increasingly leads to new findings (Brangbour et al, 2018) and observations from amateur weather stations can lead to new perspectives on local weather conditions beyond data from traditional meteorological stations (van Haren et al, 2018).…”

Section: Towards Open Weather and Climate Sciencementioning

confidence: 99%

“…In several of the studies that were presented in the conference machine learning technologies are used for data analysis and prediction (Haupt et al, 2018;Garcia-Marti et al, 2018;Bari, 2018;Schultz et al, 2018). The studies show that use of machine learning methods has added value because models are built with data beyond standard meteorological data.…”

Section: Open Softwarementioning

confidence: 99%

“…al. ;Bendoukha;Brangbour et al;Garcia-Marti et al;Haupt et al;Hut et al;Jansson et al;Pelupessy et al;Ramamurthy;Schultz et al;Stringer et al; van Haren et al;van den Oord et al, 2018). In a synthesis session we observed several developments towards open research practices and discussed challenges and opportunities.This section presents the common findings and highlights of the conference session.3.1 OPEN DATAMany studies reported in the proceedings of the conference include open data from different sources in their analyses which clearly enrich their research.…”

mentioning

confidence: 95%

“…The use of this methodology for reanalysis, i.e., generating a comprehensive and physically consistent record of how the weather is changing over time, since ensured a baseline for climate research and triggered the development of downstream climate services. Meteorologists have been using machine learning to post-process model output, blend multiple models, and optimize the weighting of those multiple models for over 20 years (Haupt et al, 2018). Neural nets were also shown to vastly speed the calculation of both incoming shortwave and outgoing longwave radiation in climate models in the 1990s (Krasnopolsky, 2013).…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Weather and Climate Science in the Digital Era

Vos¹,

Hazeleger²,

Bari³

et al. 2019

Preprint

Self Cite

View full text Add to dashboard Cite

Abstract. The need for open science has been recognized by the communities of meteorology and climate science. However, while these domains are mature in terms of applying digital technologies, these are lagging behind where the implementation of open science methodologies is concerned. In a session on <q>Weather and Climate Science in the Digital Era</q> at the 14th IEEE International eScience conference domain specialists and data and computer scientists discussed the road towards open weather and climate science. The studies presented in the conference session showed the added value of shared data, software and platforms through, for instance, combining data sets from disparate sources, increased accuracy and skill of simulations and forecasts at local scales, and improved consistency of data products. We observed that sharing data and code is important, but not sufficient to achieve open weather and climate science and that here are important issues to address. At the level of technology, the implementation of the FAIR principles to many datasets used in weather and climate science remains a challenge due to their origin, scalability, or legal barriers. Furthermore, the complexity of current software platforms limits collaboration between researchers and optimal use of open science tools and methods. The main challenges we observed, however, were non-technical and impact the system of science as a whole. There is a need for new roles and responsibilities at the interface of science and digital technology, e.g., data stewards and research software engineers. This requires the personnel portfolio of academic institutions to be more diverse, and in addition, a broader consideration of the impact of academic work, beyond publishing and teaching. Besides, new policies regarding open weather and climate science should be developed in an inclusive way to engage all stakeholders, including non-academic parties such as meteorological institutions. We acknowledge that open weather and climate science requires effort to change, but the benefits are large. As can already be observed from the studies presented in the conference it leads to much faster progress in understanding the world.

show abstract

Section: Towards Open Weather and Climate Sciencementioning

confidence: 99%

Section: Open Softwarementioning

confidence: 99%

mentioning

confidence: 95%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Weather and Climate Science in the Digital Era

Vos¹,

Hazeleger²,

Bari³

et al. 2019

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Meteorologists have been using machine learning to postprocess model output, blend multiple models, and optimize the weighting of models for over 20 years (Haupt et al, 2018). Neural nets were used in the 90s to speed up the calculation of outgoing longwave radiation in climate models (Chevallier et al, 1999) and for both short-and long-wave radiation parameterization in the National Center for Atmospheric Research (NCAR) Community Atmospheric Model (CAM) (Krasnopolsky et al, 2007).…”

Section: Introductionmentioning

confidence: 99%

Open weather and climate science in the digital era

et al. 2020

Self Cite

View full text Add to dashboard Cite

Abstract. The need for open science has been recognized by the communities of meteorology and climate science. While these domains are mature in terms of applying digital technologies, the implementation of open science methodologies is less advanced. In a session on “Weather and Climate Science in the Digital Era” at the 14th IEEE International eScience Conference domain specialists and data and computer scientists discussed the road towards open weather and climate science. Roughly 80 % of the studies presented in the conference session showed the added value of open data and software. These studies included open datasets from disparate sources in their analyses or developed tools and approaches that were made openly available to the research community. Furthermore, shared software is a prerequisite for the studies which presented systems like a model coupling framework or digital collaboration platform. Although these studies showed that sharing code and data is important, the consensus among the participants was that this is not sufficient to achieve open weather and climate science and that there are important issues to address. At the level of technology, the application of the findable, accessible, interoperable, and reusable (FAIR) principles to many datasets used in weather and climate science remains a challenge. This may be due to scalability (in the case of high-resolution climate model data, for example), legal barriers such as those encountered in using weather forecast data, or issues with heterogeneity (for example, when trying to make use of citizen data). In addition, the complexity of current software platforms often limits collaboration between researchers and the optimal use of open science tools and methods. The main challenges we observed, however, were non-technical and impact the practice of science as a whole. There is a need for new roles and responsibilities in the scientific process. People working at the interface of science and digital technology – e.g., data stewards and research software engineers – should collaborate with domain researchers to ensure the optimal use of open science tools and methods. In order to remove legal boundaries on sharing data, non-academic parties such as meteorological institutes should be allowed to act as trusted agents. Besides the creation of these new roles, novel policies regarding open weather and climate science should be developed in an inclusive way in order to engage all stakeholders. Although there is an ongoing debate on open science in the community, the individual aspects are usually discussed in isolation. Our approach in this paper takes the discourse further by focusing on “open science in weather and climate research” as a whole. We consider all aspects of open science and discuss the challenges and opportunities of recent open science developments in data, software, and hardware. We have compiled these into a list of concrete recommendations that could bring us closer to open weather and climate science. We acknowledge that the development of open weather and climate science requires effort to change, but the benefits are large. We have observed these benefits directly in the studies presented in the conference and believe that it leads to much faster progress in understanding our complex world.

show abstract