Climate indices#
The climate component offers an overview of climate indices related to hydro-meteorological hazards based on the most updated information (CMIP6). The challenge is to offer a tool that convey the complexity of climate models into statistics that are easily interpretable by non-climate experts, providign a well-rounded perspective for both space and time dimension. This is reflected on the technical side, by the challenge to reduce huge datasets into manageable pieces.
The climate component provides aggregated statistics at boundary level (country or subnational level) for a selection of:
climate-related hazards
country
time periods
The table summarises the relevant climate indices and related time scale.
Name |
Description |
Time-scale |
---|---|---|
R10mm |
Days with rainfall > 10 mm [days] |
Annual |
Rx5day |
Maximum 5-day precipitation [mm] |
Monthly |
R99p |
Extremely wet day precipitation [days] |
Monthly |
CWD |
Consecutive Wet Days [days/month] |
Annual |
CDD |
Consecutive Dry Days [days/month] |
Annual |
slr |
Sea Level Rise [m] |
Annual |
SPEI |
Standard Precipitation-EvapoT Index [-] |
Annual |
Heat |
WBGT or UTCI [°C] - bias adjusted |
Daily |
tmean |
Mean surface temperature [°C] |
Monthly |
Compared to the information directly offered by the CCKP country page, this procedure adds:
Linking relevant climate indices to natural hazard occurrance in order to estimate changes in risk drivers over the future
Standardisation of anomaly over historical variability as common metric of change
Input data#
raster data aggregated across time (20 years windows) for spatial representation
csv data aggregated across space (country boundaries) for time-series representation
includes ensemble p10 and p90 (variability across models)
Sources of CMIP6 data#
Name |
Developer |
Description |
Data format |
---|---|---|---|
World Bank |
Large selection of climate indices for both trends and extremes |
Table, geodata, charts |
|
UNDRR |
Full selection of variables, model members, periods |
Geodata |
|
IPCC |
Selection of climate variables for a range of periods and scenario |
Table, geodata, maps, charts |
Dimensions:#
SSP: SSP1/RCP2.6; SSP2/RCP4.5; SSP3/RCP7.0; SSP5/RCP8
Ensemble member: r1i1p1f1 (largest number of models available)
Ensemble range: p10, p50, p90
Period: {Historical (1981-2015)}, [Near term (2020-2039), Medium term (2040-2059), Long term (2060-2079), End of century (2080-2099)]
Time scale: Annual (R10mm, CWD, slr, SPEI); Monthly (Rxday, R99p, tmean); Daily (Heat)
Value statistic: {P10, P50, P90, SD}, [P10, P50, P90]
Script setup [WIP]#
Navigate to your working directory:
cd <Your work directory>
Run
jupyter notebook
Select CCDR-climate.ipynb and chose the indices to plot.
Processing#
Runs over one selected country and for a specific set of indices depending on selected hazard
Consider four SSP (ex RCP) scenarios (SSP1/RCP2.6; SSP2/RCP4.5; SSP3/RCP7.0; SSP5/RCP8.5)
Consider four 20-years periods (near term, medium term, long term, end of century)
Calculate median, 10th percentile (p10) and 90th percentile (p90) of standardised anomaly across models in the ensemble (more details)
Plot maps and timeseries
Exported results as csv (table) and geopackage (vector)
Output presentation#
A) Map output (spatial distribution)
Raster data aggregated across time (20 years windows)
Ensemble_mean(Period_mean(anomaly/hist_SD))
Ensemble_p50(Period_mean(anomaly/hist_SD))
ADM1-mean values from raster data
ADM1_mean(Ensemble_p50(Period_mean(anomaly/hist_SD)))
B) Chart output (time-series)
Spatial data aggregated for country ADM0 boundaries plotted as chart
Ensemble_p10(ADM0_mean(anomaly/hist_SD))
Ensemble_p50(ADM0_mean(anomaly/hist_SD))
Ensemble_p90(ADM0_mean(anomaly/hist_SD))