News that 2011 has been one of America's worst in terms of extreme weather events comes as no surprise to the thousands who have suffered freezing temperatures, tornadoes, flooding, heat waves and drought so far this year. And that's just in the US. Hurricane Irene brings to mind recent catastrophic events across the globe, including Japan's double calamity, floods and fires in Australia, and the crippling drought in the Horn of Africa.
While it's difficult to link single events such as hurricanes directly to climate change, scientists have warned that the trend of extreme weather is likely to continue. And while dramatic events get more exposure, it's the subtler, long-term shifts in regional climates that will have an equally, if not more, devastating effect on inhabitants.
It's these regional pattern changes that are now being extensively researched, as scientists try to gain a clearer picture of the future climate. With last year reported as the warmest on record, humans are becoming increasingly vulnerable to climate change – and gathering data to better determine how, where, and when this will happen could prove invaluable.
Given the southern half of the continent is likely to experience particularly severe impacts, scientists are working to produce climate projections for this region – and using hi-tech computing to do it. The CSIR, for example, is working on a climate simulation experiment for Africa, using a computer cluster at the Johannesburg-based Centre for High Performance Computing (CHPC). The distributed computing environment is something CSIR researcher Francois Engelbrecht says is essential for climate science in SA.“Supercomputing is extremely important because climate modelling relies on very complex mathematical models to do calculations. It involves the numerical solution of big equation sets and for this activity we use a type of computing called parallel processing.”
This environment sees tens to hundreds of processors working simultaneously on the same calculations, speeding up project times significantly. “For example, when we were working on a climate simulation for a period of 140 years, six simulations took three months – but if it weren't for the supercomputers, it would have taken years.”
Engelbrecht adds that recent experiments provided several important findings, the first being the rapid rise in temperature in southern Africa over the next century.
“If greenhouse gases continue to grow at current rates, the Southern African interior could experience temperature increases of four degrees to six degrees Celsius. This would have huge impacts,” says Engelbrecht, including lower crop yields, as very high temperatures are damaging for most crops.“The southern African region will become generally drier while at the same time the east African region will become wetter.” The area likely to be particularly vulnerable to climate change in SA is the south-west Cape, because it's used to cool and wet winters, and there are strong signals that the area will become hotter and drier, he adds.
This could spell trouble for the Cape wine industry, as high temperatures generally decrease the quality of wines, notes Engelbrecht.
Other dangers include increased pressure on water resources, the transmission zone for malaria spreading further southward, and plant ecosystems such as Fynbos and succulent Karoo coming under threat. “Vegetation adapts to the climate over thousands of years, and now you have significant changes happening in a space of decades.”
According to Engelbrecht, this is the most extensive experiment of its kind performed to date, thanks to the support of the CHPC. He says the international climate science community has realised Africa is particularly vulnerable to climate-related effects, and has stepped up research efforts on the continent.
The Sarva site notes that a project is under way to better the understanding of Southern African Development Community member states on the impacts and risks of climate change. A new climate risk handbook, to be completed in the second half of this year, will serve as a basic reference guide to those engaged in risk and vulnerability research.
Several organisations are engaged in the project, including the CSIR's Climate Studies and Modelling Group, Kulima Development Solutions, and the Climate Systems Analysis Group at the University of Cape Town. Engelbrecht says the climate modelling community in SA collaborates closely on a number of projects, because it's quite a small group.
Collaborating also helps researchers draw conclusions when robust climate signals are in agreement – such as the fact that southern Africa will experience rapid rises in general temperature.
“It's not a perfect science, but we've gotten to the point where we can pick up large-scale trends with a very important message, and there's enough confidence in that message to take action.”
Six simulations took three months – but if it weren't for the supercomputers, it would have taken years.
One of the resulting projects is the Climate CoLab, an open online forum inviting people to suggest and discuss actions that address climate change and the green economy. The site engages scientists, policy-makers, business people and individuals to create an evolving collection of proposals, which can be developed through collaborative input from other members of the community.
Climate CoLab project manager Robert Laubacher notes that new types of systems use the collective intelligence of people from around the world to create huge crowd-sourced databases, such as Wikipedia.
“The Climate CoLab is a system we're trying to build that embodies the concept of collective intelligence. We saw the emergence of very large pools of data which can seem quite daunting, so we looked for a new mechanism to bring people together across geographies.”
The site's main project is an annual contest, aimed at sourcing and selecting innovative and viable climate change initiatives, which are chosen through a process of community input, scientific modelling, expert judging and voting.
Once all the proposals have been submitted, computerised simulation models are used to project the environmental and economic outcomes of the proposed actions (such as sea level rise or mitigation costs). In the next stage, a group of climate scientists and policy experts reviews and selects the finalists. These teams are given time to improve their projects, and Climate CoLab members are then invited to vote for the most promising proposals. Winners are chosen based on voting and another judging assessment. This year's winning proposals will be included in briefings being planned for policy-makers at the United Nations and US Congress.
Laubacher says because climate change is such a complex problem, structuring activities on the site proved challenging. “It's different to an encyclopaedia that just consists of separate articles linked together; there's a lot of interaction in the system, with highly complex parts. We were wrestling with that in the first few attempts at structuring activities.”
He explains that fairly broad questions are used for the contest structure, but that organisers hope to include a more detailed breaking down of problems next year.
“We'd like to be able to flesh out topics in each problem domain, such as what to do at global, national, local, and household level, with a number of options under each level.
“Lots of citizens who have specific expertise, and enough of them working on a domain, could be very powerful, but we're not quite there yet.”
A key part of assessing global proposals is determining their potential effects through modelling systems. “One of the things we thought was interesting in the domain of climate change is that there are lots of models that project what will happen if we do certain things, in areas such as temperature or sea levels,” says Laubacher.
Another piece of the puzzle is connecting these ideas to the policy world – such as when can they be carried through or become law.
“Some policy-makers are very intrigued by the possibility of getting the broad voice of citizens into the conversation, while others like to keep the conversation among themselves. We've created ties with national government constituencies, and usually try to identify people who are tech-savvy and more inclined towards this type of approach, such as supporters of e-government.“
He says the site received about 22 proposals in the global contest, and that organisers tried to encourage entries by getting experts to give input before the final submission date. “Many are still early-phase drafts, but the contest's certainly gotten attention, with visitors to the site from almost 150 countries.”
Laubacher says it's often tricky to balance inputs coming from the public and those coming from the expert panel. “On the one hand, there are lots of concerned citizens with valuable ideas, and on the other, you have experts who have spent years studying these issues. We've attracted plenty of ideas in specific areas, but it's been difficult to get a broadly integrated vision of how the economy could evolve going forward.
“To a great extent, if we can break down a problem into smaller parts, we could take better advantage of citizen expertise.”
With social networks giving people more avenues for speaking out and taking action, Laubacher says grass roots efforts could potentially create movement at policy levels. “We'd love it if this kind of approach could become perceived as a valuable input into traditional policy processes, and eventually become a core part of it.”
Dealing with catastropheWhile some projects are looking into the future, others are finding ways to deal with disasters that do happen. Carnegie Mellon University's Silicon Valley (CMUSV) campus, for example, recently partnered with Nasa for an emergency briefing and response centre.
According to CMUSV, emergency managers and response teams often need to find and assess damage over huge areas that can stretch over thousands of kilometres. Researchers at the university are developing hi-tech tools to help speed up disaster relief efforts and recovery.
The university's Next-Generation Emergency Operations Centre is being used as a mission control centre for Nasa's Wildfire Research and Applications Partnership programme, which is running flights during September and October. During this time, Nasa and CMUSV teams will be on 24-hour call to fly over developing wildfires, collecting data which they can instantly send back to response commanders on the ground.
“Although this work has focused on wildfires, this same technology will be valuable after an earthquake, a tsunami, an oil spill or any other disaster that covers a large area,” said Art Botterell, CMUSV Next-Gen EOC consultant, in a release.
“We've had aerial imagery for a long time, but never a capacity like this for putting together a detailed wide-area operating picture in real-time,” he added. “Advanced computer image analysis promises to help us detect things like natural gas leaks that simple 'eyeball' review of the images can easily miss.”
The partnership forms part of CMUSV's Disaster Management Initiative (DMI), which was formed in 2009 to provide open, interoperable technical solutions for emergencies that stretch across a number of regions.
Eating an elephant
If we can break down a problem into smaller parts, we could take better advantage of citizen expertise.
“As people at the site of a disaster send text messages and post to sites such as Twitter and Facebook, the volunteers in groups such as CrisisCommons are geo-locating messages, which aid decision-makers in assessing a situation and triage resources without being on site.
“There is great potential for social media to provide real-time status of infrastructure that could be used, for example, to alert first responders in advance if the route to a location is blocked by a collapsed bridge,” says CMUSV.
It adds that several challenges can hinder the fast transfer of information, such as language barriers or the volume of the Twitter stream. For emergency managers, accuracy of information is a key issue. The DMI researchers hope to find ways to help identify accurate reports, correct inaccurate ones, and use the power of collaborative networks to get the right information to the right people, when it matters.
Our comments policy does not allow anonymous postings. Read the policy here