This application is written in Python, with the help of the ReportLab library (the web part is powered by the flask framework). The code is open-source and is available here: framagit.org. The ability of this application to reproduce several figures (independently) published by the IPCC was carefully tested. However, this is not a product of the IPCC (which is not responsible for projects or researches).
If it does not work as expected
We are sorry to hear that you have encountered a problem. We will do everything we can to help. Please have a look at 'It did not work! Why?' in the tutorial. Do not hesitate to provide feedback.Development history
Although the web application and layout details were improved over time, there is no change affecting the translation of the risk levels provided as input to colours in the burning embers. The changes are recorded on framagit.org (see Changelog). A short summary is provided below (from recent to old).
Recent news?
When an update is planned, it might be announced on climrisk.org.
2023/04: Support for detailed risk transitions, simplifications and corrections
The update provides support for complex descriptions of transitions, which may involve one or both of the following:
- risk transition ranges assessed for more than one confidence level (e.g. "it is likely that it will occur in this narrow range, but very likely that it will occur within that broader range").
- more information about the hazard level that corresponds to a given risk level inside a "risk transition". The standard is min, (optional) median, max; the update adds the possibility to provide "percentiles", which are intermediate levels of risk within a transition: p25, p75, etc. (p50 would be the median).
The update simplifies the documentation by referring to the former "basic (file) format" as the standard
format. The former "full-flex" format was probably used in a single case and is kept for legacy file(s).
The connecting lines showing changes between embers will no longer be drawn for risk levels absent from an
ember.
When non-numeric data is erroneously provided for an hazard level (wighin a transition), it will be converted
to numeric with a warning. This typically results from entering "." instead of "," as the decimal separator in
Excel, with non-English languages. Such erroneous lines were previously ignored: there might thus be a change in
behavior (with a warning).
(version 1.7.1)
2021/09: Support for 'overlapping transitions', corrections for technical issues
It is now possible to define 'overlapping transitions', i.e. generate embers with 'uncertainty ranges' which span
across each other (e.g. it might be that risk is above or below 'high' at a given level).
Technical changes enable files to be opened in recent versions of Adobe Illustrator (...).
(version 1.6)
2021/01: Additional layout options
Optional secondary axis (on the right, typically with a different scale), minor tick marks
(unlabelled ticks between the main axis ticks), tick marks instead of horizontal grid lines, legend for confidence
levels (experimental).
(version 1.5)
2020/12: Improvements, full documentation of the parameters, bug fix for specific cases
File format download options (PDF/PNG/JPEG) are added. Cases where the vertical scale ('hazard') has limits significantly
different from the usual "0 to +5°C", such as for CO2 concentrations in ppm, are now much better handled (bug fix).
Coding improvements provide more clarity for future development. All parameters are documented in a way
that is fully consistent with how they work (the same file provides online documentation and default values).
(version 1.4)
2020/08: Median value in the transitions and better handling of input file issues
An optinal median, or 'midpoint' between the start and end of each transition, is added in the 'Basic format'. This results in the 'standard template' provided in Tutorial. Input files containing missing data, errors, or other unexpected characteristics are better handled.
2019/11 - 2020/05: Initial development
This software was created by Philippe Marbaix at the end of 2019
(based on an earlier 'AR5-related' back-of-the-envelope version).
The first objective was to produce figure 3 of Zommers et al. (2020;
Burning Embers: Towards more transparent and robust climate change risk assessments.
Nature Reviews Earth & Environment. doi.org/10/gg985p).
Help is welcome to further improve the application. All contributions will be recognised :-). For more information, please e-mail philippe.marbaix@uclouvain.be.