The Science of Doom

In VI – Australia CanESM2, CSIRO, Miroc and MRI compared vs history we looked at how each model thought rainfall had changed in Australia over about 100 years, and we compared that to observations. We did this for annual rainfall, also for Australian summer (Dec, Jan, Feb) and Australian winter (Jun, Jul, Aug).

Here we will look at two of the four emissions scenarios. We compare 2081-2100 vs 1979-2005.

Note that we are not comparing the end of the 21st century from the model with observations at the end of the 20th century. That produces much different results – the model’s view of recent history doesn’t match observations very well. We are comparing the model future with the model past. So we are asking the model to say how it sees rainfall changing as a result of different amounts of CO2 being emitted.

The two scenarios are:

• RCP4.5 – with current trends continuing we are something like RCP6. I think of RCP4.5 as being “what we are doing now” but with some substantial reductions in CO2 emissions. But it’s nothing like RCP2.6, which is more “project Greta” where emissions basically stop in a decade
• RCP8.5 – extreme CO2 emissions. Often described as “business as usual” perhaps to get people’s attention. Think – most of Africa moving out of abject poverty, not passing through the demographic transition (so population going very high) and burning coal like crazy with the efficiency of 19th century Europe.

Each pair of graphs is future RCP4.5 as % of recent past, and RCP8.5 as % of recent past. The four models, clockwise from top left – MPI (Germany), Miroc (Japan), CSIRO (Australia) and CAN (Canada):

Figure 1 – Click to expand

And now the same, but only looking at Australian summer, DJF:

Figure 2 – Click to expand

Depending on which model you like, things could be really bad, or really good, or about the same with “climate change”.

Note that the color scale I’m using here is the same as the last article, but different from all the earlier articles, the % range is from 50% to 150% (rather than 0% to 200%).

References

An overview of CMIP5 and the experiment design, Taylor, Stouffer & Meehl, AMS (2012)

GPCP data provided by the NOAA/OAR/ESRL PSL, Boulder, Colorado, USA, from their Web site at https://psl.noaa.gov/

GPCC data provided from https://psl.noaa.gov/data/gridded/data.gpcc.html

CMIP5 data provided by the portal at https://esgf-data.dkrz.de/search/cmip5-dkrz/

Read Full Post »

In V – CanESM2, CSIRO, Miroc and MRI compared we compared four models among themselves for two future scenarios of CO2 emissions, and also the four models compared with historical observations.

Here we zero in on Australia. Let’s compare all months 1979-2005, i.e. recent history with around 100 years before that, all months 1891-1910 (note 1).

This first figure is a % comparison. Each map is annual data: average 1979-2005 % of average 1891-1910. Note that the color scale I’m using here is different from previous articles, the % range is from 50% to 150% (rather than 0% to 200%).

The left-most map is observations, GPCC, and on the right the four different models. Each of the four maps is one model, 1979-2005 as a % of that model for 1891-1910 – clockwise from top left, MPI, MIROC, CSIRO, CanESM2 (note 2):

Figure 1 – Click to expand

So we are seeing how well the models compare among themselves, and with observations, for a century or so change. All of the models are run with the identical set of conditions (the best estimate of forcings like CO2, aerosols, etc) – that’s what CMIP5 is all about.

This second graphic is % comparison over Australian summer: December, January, February (DJF). It is otherwise exactly the same as the figure 1:

Figure 2 – Click to expand

The annual model comparisons look “better” than the summer (DJF) comparisons.

With the DJF comparisons, Australian summer observations across a century have the western half of Australia wetter, and coastal Queensland (that’s the right edge from halfway up) drier. Also some inland NSW regions drier.

MPI and CSIRO show the western edge drier. Miroc and CAN show the western edge wetter. CSIRO has the Adelaide region and west much drier, observations show much wetter, CAN and MPI show this area a little wetter while Miroc has it about the same.

It’s difficult to claim the summer model comparisons demonstrate any insight – given that we can check them against observations. And overall, these four models don’t demonstrate any particular biases, i.e., they don’t all agree with each other against the observations. Apart from inland western Australia where they fail to predict the much higher rainfall seen in observations.

Place yourself back in 1900. You have these models, how useful are they for predicting 100 years ahead what would happen to summer rainfall?

References

An overview of CMIP5 and the experiment design, Taylor, Stouffer & Meehl, AMS (2012)

GPCP data provided by the NOAA/OAR/ESRL PSL, Boulder, Colorado, USA, from their Web site at https://psl.noaa.gov/

GPCC data provided from https://psl.noaa.gov/data/gridded/data.gpcc.html

CMIP5 data provided by the portal at https://esgf-data.dkrz.de/search/cmip5-dkrz/

Notes

Note 1: The choice of dates is constrained by:

• 1891 being the start of the GPCC observational dataset
• 1979 being the start of the satellite era
• 2005 being the end date that this class of models ran to for their “historical” simulation – CMIP5 historical simulations were from 1850-2005

As a result, lots of comparisons in climate papers involve 1979-2005, so even though we aren’t using satellite data here, I have been using that 27-year period.

Note 2: Each model output is the median of all of the simulations

Read Full Post »

In the last article we looked at a comparison between Miroc (Japanese climate mode) and MPI (German climate model). See that article for more details.

Now we add CanESM2 and CSIRO-Mk3-6-0 to the comparison.

CanESM2 is a Canadian climate model, with an ESM component – this is an earth system model, basically it means that CO2 emissions are explicitly controlled, but not the atmospheric CO2 concentration (so the model simulates aspects of the carbon cycle). Their model has 5 historical simulations and 5 each each of three RCPs (skipping RCP6 like many other CMIP5 contributors)

CSIRO-Mk3-6-0 is an Australian model. Their model has 3 historical simulations and 10 each of the four RCPs.

As in the previous article, MPI, Miroc, CAN and CSIRO for RCP4.5 for 2081-2100. Each graphic – the median of all of the simulations as % of the median of that model’s historical 1979-2005 simulations:

Figure 1 – MPI, Miroc, CAN & CSIRO for RCP4.5 (%) – Click to expand

And for RCP8.5 for 2081-2100

Figure 2 – MPI, Miroc, CAN & CSIRO for RCP8.5 (%) – Click to expand

And comparisons of each models’ historical runs (the median of multiple runs): % of observations (GPCC) over 1979-2005. So blue means the model over-estimates actual rainfall, whereas red means the model under-estimates:

Figure 3 – MPI, Miroc, CAN & CSIRO historical runs compared with GPCC over the same 1979-2005 period – Click to expand

Clearly a strong consensus.

Read Full Post »

In Models and Rainfall – III – MPI Seasonal and Models and Rainfall – II – MPI we looked at one model, MPI from Germany, from a variety of perspectives.

In this article we’ll look at another model that took part in the last Climate Model Intercomparison Project (CMIP5) – Miroc5 from Japan and compare it with MPI.

A reminder from an earlier article – the scenarios (Representative Concentration Pathways) in brief (and see van Vuuren reference below):

• rcp2.6 – large reductions in CO2 emissions within a short space of time. Conceptual model – shutting off the world’s power stations, and no burning of fossil fuels, by 2030. Think: Project Greta
• rcp 4.5 – substantial improvements in reducing CO2 emissions
• rcp 6 – roughly where we will be in 2100 based on current trends
• rcp 8.5 – extreme CO2 emissions, often misleadingly cited as “business as usual” (see Opinions and Perspectives – 3 – How much CO2 will there be? And Activists in Disguise and Opinions and Perspectives – 3.5 – Follow up to “How much CO2 will there be?”)

Miroc5 (just called Miroc in the rest of the article) did five simulations of historical and three simulations of each RCP through to 2100.

The first graphic has five maps: first, the median Miroc simulation of 1979-2005, followed by simulations of 2081-2100 for rcp2.6 to rcp8.5 (each one is the median of the three simulations):

Figure 1 – Miroc simulations of historical 1979-2005 and the 4 RCPs in 2081-2100 – Click to expand

The % change of the median Miroc simulation for each scenario from the median historical simulation:

We can see a consistent theme through increasing CO2 concentrations.

Figure 2 – Miroc simulations for RCPs 2081-2100 as % of Miroc historical 1979-2005 – Click to expand

As the previous figure, but difference (future – historical):

Figure 3 – Miroc simulations for RCPs 2081-2100 less Miroc historical 1979-2005 – Click to expand

Side by Side Comparisons of MPI and Miroc Predictions

And now some comparisons side by side. On the left MPI, on the right Miroc. Both are comparing RCP4.5 as a percentage of their own historical simulation (and both are the medians of the simulations):

Figure 4 – MPI compared with Miroc for RCP4.5 (%) – Click to expand

I think seeing the future less historical (as a difference rather than %) is also useful – in areas with very low rain the % difference can appear extreme even though the impact is very low. Overall, % graphs are more useful – if you live in an area with say 20mm of rainfall per month on average then -10mm might not show up very well on a difference chart, but it can be critical. But for reference, the difference:

Figure 5 – MPI compared with Miroc for RCP4.5 (difference) – Click to expand

Now the same two graphs for RCP8.5. On the left MPI, on the right Miroc. % of their historical simulation in each case:

Figure 6 – MPI compared with Miroc for RCP8.5 (%) – Click to expand

And now difference (future less historical) in each case:

Figure 7 – MPI compared with Miroc for RCP8.5 (difference) – Click to expand

Side by Side Comparisons of Models vs Observations

In Part II we saw some comparisons of the MPI model with GPCC observations, both over the same 1979-2005 time period. Here is MPI (left) and MIROC (right) each as a % of GPCC:

Figure 8 – MPI compared with Miroc for GPCC observations (%) – Click to expand

It’s clear that different models, at least for now MPI and Miroc, have significant differences between them.

References

An overview of CMIP5 and the experiment design, Taylor, Stouffer & Meehl, AMS (2012)

GPCP data provided by the NOAA/OAR/ESRL PSL, Boulder, Colorado, USA, from their Web site at https://psl.noaa.gov/

GPCC data provided from https://psl.noaa.gov/data/gridded/data.gpcc.html

CMIP5 data provided by the portal at https://esgf-data.dkrz.de/search/cmip5-dkrz/

The representative concentration pathways: an overview, van Vuuren et al, Climatic Change (2011)

Read Full Post »