Preface to the thematic issue on Environmental Data Science. Applications to air quality and water cycle
This special issue harvested a set of papers that provide a useful perspective of how Data Science tackles nonlinear, spatio-temporal environmental phenomena using a variety of data sources (numeric variables, qualitative, ordinal, time series, videos, smart sensor data) for both descriptive and predictive processes. The included papers employ a variety of techniques, ranging from classical statistical methods to innovative pre-processing methods or hybrid machine learning methods to extract knowledge from data. The selection of papers for this issue was done with a rigorous blind peer-review process and high rate of rejection. The issue originally received 29 submissions, and the 8 papers most favorably evaluated by reviewers were selected. The reviewing process involved more than 170 reviewers. The selected papers provide a nice overview of research conducted in multiple countries, including international collaborations and some collaboration between academia and corporations.
K. Gibert, J. S.Horsburgh, I. N.Athanasiadis, G. Holmes, Preface to the thematic issue on Environmental Data Science. Applications to air quality and water cycle, Environmental Modelling and Software, 106:1-3, 2018, doi:10.1016/j.envsoft.2018.03.020.
You might also enjoy (View all publications)
- A weakly supervised framework for high-resolution crop yield forecasts
- Learning latent representations for operational nitrogen response rate prediction
- Simulation-assisted machine learning for operational digital twins