Event

DailyMeteo.org/2014 Conference

Location: 

University of Belgrade - Faculty of Civil Engineering Belgrade
Serbia

Event date: 

Monday, June 23, 2014 - 09:00 to Friday, June 27, 2014 - 17:00


Spatial and spatio-temporal modelling of meteorological and climatic variables using Open Source software (R + OSGeo)

International conference and workshop: three days of software tutorials (DailyMeteo.org WORKSHOPS) + two days of conference with keynote speakers. Register for this event.

 

Aims and scope:

This workshop aims at forming an international consortium to support production of an open archive of meteorological images at high spatial (1 km) and temporal (1 day) resolution (the WorldDailyMeteo initiative). We look at implementing the state of the art spatio-temporal geostatistics (with the focus on algorithms implemented in the Open Source Software: R, GRASS GIS, SAGA GIS) to produce a public repository of meteorological images (as an update to the existing global climatic archives e.g. WorldClim). We invite all international and national teams that work on mapping climatic and meteorological variables to come to Belgrade and present their work, participate in the discussion forums and consider joining the WorldDailyMeteo initiative.

The workshop will consist of a 3-day training session (tutorials) on spatio-temporal interpolation of meteo data in R (packages: raster, gstat, spacetime, meteo, plotGoogleMaps, plotKML) lead by the original package developers, and a 2-day conference with keynotes.

Themes:

  • Spatio-temporal interpolation of daily temperature, rainfall, meteorological conditions
  • Global climatic and meteorological data sets
  • Combining geostatistics and remote sensing
  • Time-series analysis of meteo data
  • Automated detection of climatic change
  • Sampling optimization for monitoring climate change
  • Multivariate geostatistics for climatology
  • New software for spatial analysis in meteorology

Conference programme:

Day to day programme is available here...

Keynote speakers:

Edzer Pebesma

"Spatial and temporal support of meteorological observations and predictions"

Prof. dr. Edzer Pebesma, University of Münster, Germany

For spatial or spatio-temporal prediction, with the exception of cross validation one is rarely interested in predicting the observation process, but instead a smooth form that filters measurement error, or aggregates over space and/or time. The talk will address this issue, its advantages, challenges, practical examples and implementation details. It will also address the problem of interpolating yearly aggregated (e.g. mean, maximum) values from yearly aggregated observations, or from e.g. interpolated daily values.

Edzer Pebesma is director of the institute for Geoinformatics of the University of Münster, where amongst others he teaches geostatistics and modelling of spatio-temporal processes. He is author and maintainer of several key spatial packages for R, including sp, gstat and spacetime. Together with Roger Bivand and Virgilio Gomez-Rubio, he co-authors the popular Springer volume “Applied Spatial Data Analysis with R”. He is associate editor for Spatial Statistics, Computers & Geosciences, and the Journal of Statistical Software.

"An operational R-based interpolation facility for climate and meteo data"

dr. Raymond Sluiter, KNMI, the Netherlands

This lecture addresses the latest developments of KNMI’s GeoSpatial Interpolation Environment (GSIE), the interpolation techniques, the datasets and data dissemination through the KNMI Data Centre (KDC).  Topics of interest: R, gstat, automap, OGC web services, ADAGUC, INSPIRE, metadata, data quality, satellite data assimilation and open data.

Raymond Sluiter is senior researcher at the Royal Netherlands Meteorological Institute (KNMI). He coordinates the implementation and use of GIS within KNMI. His research is focused on geospatial data infrastructures, both national and international, and on spatialization of climatological and meteorological variables. Being national delegate at ESA’s Data Operations Scientific and Technical Advisory Group (DOSTAG) he is well aware of the state of art of earth observation.

"Statistical modelling of space-time variability"

dr. Gerard B.M. Heuvelink, Wageningen University

This talk will review the main statistical approaches, such as the multivariate approach where correlations between spatial variables at different times is treated using a cokriging model, the 3D approach where time is an extra dimension and differences between time and space are modelled through trends and anisotropies, and the state-space approach that employs mechanistic knowledge and uses techniques such as the space-time Kalman filter.

Gerard Heuvelink is associate professor in Geostatistics with the Soil Geography and Landscape group of Wageningen University  and senior researcher Pedometrics and Digital Soil Mapping with ISRIC — World Soil Information. Gerard is worldwide recognised as a leading scientist in pedometrics and spatial uncertainty analysis, mainly through his publications  and active involvement in various pedometrics and accuracy-related conference series. His research areas are geostatistics, pedometrics, spatial uncertainty analysis, sampling design optimisation, spatial aggregation and disaggregation, and space-time kriging. He is associate editor of the European Journal of Soil Science and Spatial Statistics, and editorial board member of Geoderma, Environmental and Ecological Statistics, International Journal of Applied Earth Observation and Geoinformation and Geographical Analysis.

"Production of climate maps: operational issues and challenges"

Mojca Dolinar, Environmental Agency of Republic of Slovenia

When calculating climatologies in regular grids geostatistical methods prove to be very powerful. However, in operational mode, we are faced by many challenges as the consequence of imperfect reality, like measurement errors, poor density and representativeness of measurement network... Additionally, every climatological variable has its own peculiarity, which should be solved during the spatialisation process. In this lecture some specific problems (like temperature inversion, the lack of high mountains measurements…) will be addressed and some possible solutions described in the case of highly variable Slovenian climate. Additionally, the operational production of Slovenian climate maps will be described: database and interpolation techniques, auxiliary variables, postprocesing, user demands...

Mojca Dolinar is a head of Climatology Department at Slovenian Environment Agency. She works on many fields of Climatology, however, her research work is mainly focused on spatialisation of climatological variables and local climate change issues. For many years she's been co-convener of Spatial Climatology Session of EMS Annual Meetings and guest editor of Advances in Science and Research.

 

"Modeling resilience to climate change in space and time"

Prof. dr. Slobodan P. Simonovic, University of Western Ontario

Change of climatic conditions is linked to increasing number of catastrophic weather caused disaster events across the world (floods, droughts, hurricanes, and tornadoes) and led to loss of thousands of human lives and shattered the global economic growth. This talk with present an innovative climate change impact research that focuses on the concept of resilience – measure that integrate vulnerability and adaptive capacity of a complex system. The presentation will introduce a framework for mapping resilience in time and space (dynamic maps) to inform decision making processes. 

Dr. Slobodan P. Simonovic is Professor at the Civil and Environmental Engineering Department of the University of Western Ontario and Research Chair with the Institute for Catastrophic Loss Reduction. Professor  Simonovic is internationally recognized expert with interest in the application of systems approach to, and development of the decision support tools for, management of complex water and environmental systems.  Most of his work is related to the application of computerized simulation, optimization and multi-objective analysis tools in deterministic, probabilistic and fuzzy form. The second focus area of Dr. Simonovic’s research includes risk management. He is developing, and implementing, probabilistic and fuzzy set methods for water resources risk management. He has undertaken applied research projects that integrate the mathematical modeling, spatial and temporal data-base management and intelligent interface development into decision support tools for water resources decision makers. Most of his research is being conducted through the Facility for Intelligent Decision Support (FIDS) at the University of Western Ontario. He has received a number of awards for excellence in teaching, research and outreach. He has published over 400 professional publications and three major textbooks. He has been inducted to the Canadian Academy of Engeering in June of 2013.

 

Webinars:

"New global climate surfaces, and the trade-off between space and time"

dr. Robert Hijmans, University of California, Davis

New global climate surfaces (WorldClim v2) will be presented. This new data set includes additional variables; and the quality of the data has much improved by adding additional co-variates in the interpolation. Costs and benefits of the use of daily data data and complex models vs more aggregated climate data and simple data will be discussed.

Robert is an associate professor at the UC Davis California. Robert is the author of multiple R packages including ‘raster’ for analyzing raster data and ‘dismo’ for species distribution modelling. Robert leads the developement of the WorldClim.org global climate interpolated data at 1 km resolution.


Conference organizers (hosts):

  1. Milan Kilibarda (conference secretary, Faculty of Civil Engineering, University of Belgrade)
 

 

Venue:

University of Belgrade logo

University of Belgrade
Faculty of Mechanical Engineering / Faculty of Civil Engineering

Conference hall

Cafeteria

Terrace (on the roof!)


Pre-conference tutorials (23.-25. June, 2014):

Edzer Pebesma
  Edzer Pebesma: spacetime / gstat package tutorials (spatial and spatio-temporal statistics with R)
Milan Kilibarda   Milan KilibardaplotGoogleMaps / meteo package tutorials
Tomislav Hengl
  Tomislav Hengl: plotKML package tutorial (visualization of spatial and spatio-temporal data from R to Google Earth)
test
  Benedikt Grälergstat package tutorial (choosing and fitting spatio-temporal models for meteorological data)

 


 

Important dates:

  • March 30th 2014 — abstract submission / registration deadline

  • May 1th 2014  draft programme and invitation letters

  • May 15th 2014  deadline early birds

  • May 30th 2014  final programme

  • June 23rd 2014  start of the conference (deadline registration)

 Registration is following 1st register for the website, then for the event.


Registration fees:

Early birds PhD students and MSc students (conference):*

145 EUR

Early birds full registration fee (conference):

220 EUR

Registration fee (conference) after May 15th 2014:

285 EUR

Registration fee (3-day training courses) PhD/MSc students:*

95 EUR

Registration fee (3-day training courses) regular:

152 EUR

Early birds conference and training courses PhD/MSc students:*

195 EUR

Early birds full registration fee conference and training courses:

345 EUR

Full registration fee conference and training courses after May 15th 2014:

410 EUR

* Proof of enrolment in a MSc / PhD programme must be provided. 


Submission of extended abstracts and full papers (*New):

Submission for conference proceedings is possible in form of extended abstract or full paper. For extended abstract the maximum length of a manuscript is 3 pages in provided template. For full paper the maximum length of a manuscript is 6 pages in provided template. Deadline for submission is 20th of May 2014.

DOC TEMPLATE FOR DAILYMETEO/2014.

 A number of selected extended abstracts will be invited for submission for the special issue on "Spatial and spatio-temporal models for interpolating climatic and meteorological data" (Spatial Statistics journal; associate editors: Hengl, Pebesma and Hijmans).

  • Submissions open: 1st of July 2014

  • Submissions close: 15th of October 2014  1st of November 2015

  • Acceptance deadline (closing of the special issue): 1st of March 2015

Submission guidelines:

  1. Download the Elsevier article template (e.g. LaTeX template from http://www.elsevier.com/author-schemas/latex-instructions);

  2. Study the themes of interest (http://dailymeteo.org/2014#toc-themes-9HSRW43W);

  3. This special issue is about methods for interpolating climatic and meteo data, but also about using Open Source software to achieve this. Consider using a combination of R and/or Python and LaTeX code to produce papers that include both formulas and code snippets.

  4. Once you have managed to compile a PDF of your article for peer-review, visit the SPASTA editorial system at http://ees.elsevier.com/spasta/default.asp, register a new author account (if required) and then login and upload your article.

  5. During the article submission, you need to select the right article type "SI:Dailymeteo.org/2014" when you reach the "article type" step in the submission process, so that their papers will be routed together for the special issue into the right channel, not get mixed with other SI papers, or regular papers in the system.

Arriving to Belgrade and accommodation

 ... more information

Event Calendar

S M T W T F S
 
 
 
 
 
 
1
 
2
 
3
 
4
 
5
 
6
 
7
 
8
 
9
 
10
 
11
 
12
 
13
 
14
 
15
 
16
 
17
 
18
 
19
 
20
 
21
 
22
 
23
 
24
 
25
 
26
 
27
 
28
 
29
 
30
 
 
 
 
 
 
 

Meteo package

Recent comments

No comments available.