What are the local consequences of a continued global warming? And what kind of future climate can you expect for you children? Do we expect more extreme events, and will a global warming affect the statistics of storms? Another question is how the local changes matters for local communities and the ecosystem.
It may be contrary to most people’s impression. We have a clearer picture of future climate changes on a global scale than of the local consequences associated with a global warming. And we know why.
It’s reasonably predictable that the global mean temperature will increase according to the models’ climate sensitivity. But it’s harder to answer the question where the storm tracks be in the future, or what will happen to El Nino Southern Oscillation (ENSO) in a warmer world. And how will the heat be distributed in the oceans?
Climate models, like all models, are designed with some limitations. Their objective is to describe features that are of key interest for a given task, and not reproduce all the details seen in nature. They typically capture the coarse aspects of large-scale weather phenomena, which often implies that they may be somewhat displaced in terms of their real location.
Furthermore, they reproduce much of the natural variability seen in the real world, but also indicate that it is impossible to predict their exact future path beyond the time horizon of weather forecasting (see previous comment on Deser et al. 2012). Nevertheless, we can get some idea of the range of plausible outcomes by making many model predictions (ensemble runs, explained as “Monte-Carlo simulations” in the excellent BBC documentary “Climate Change by Numbers“).
The local climate can be regarded as the same as weather statistics, providing a picture of expected ranges and occurrences of different atmospheric phenomena. A climate change then implies a change in the weather statistics, with changes in frequencies and ranges. Some weather phenomena are dangerous, and hence a change in their occurrence means there will be a change in weather-related risks.
In other words, we know that Earth’s climate is changing, but we do not know exactly what the consequences will be locally where you live. However, we can make some estimate of weather-related risks. The problem is to provide a bridge between the scientific knowledge and information that is directly relevant and tailored for decision-making.
The local dimension is important for climate change adaptation and for many decision-makers, and it is important to figure out how the climate-related risks may change on specific locations in order to be prepared. For this reason, the so-called global framework for climate services (GFCS) was established by the World Meteorological Organisation (WMO).
There are also initiatives that try to enhance our understanding of regional and local climate change, such as the COoRdinated Downscaling EXperiment (CORDEX) under the World Climate Research Programme (WCRP).
Now, the downscaling efforts are getting a renewed vigour within the IPCC, and Brazil’s INPE recently hosted a workshop on regional climate projections and their use in impacts and risk analysis studies (link).
One key question is how to make best use of our knowledge and information in decision making and climate change adaptation. How to make decisions based on climate science? To achieve this objective we need to ask who are using the information. And what do people actually need?
Not surprising, one take-home message from the workshop was that dialogue between decision-makers and scientists is important. And it is important that the scientists and users of the information (such as decision-makers or scientists from other disciplines) understand the limitations of the data and what are the consequences of climate change.
Some of the hottest debated concepts at the IPCC workshop involved “model bias” and “bias correction“, but not all who need climatological data may understand their meaning. Not that they are stupid, but many discussions on this topic are cryptic for those outside the climate research community.
The obstacles associated with mutual understanding across different scientific disciplines also surfaced during a more recent workshop on biodiversity and climate change in Bogota, Colombia (link), organised by Alcue Net and CORDEX and hosted by Colciencias. One conclusion from this meeting was that the mutual understanding across scientific disciplines may improve through working closely together over time. The experts must come out of their comfort zone.
Furthermore, data sharing facilitates better understanding, however, it’s important to document the limitation of model results and distinguish between what is model results and what is observations. Data need to be accompanied with unambiguous and standardised metadata.
In other words, there’s a need for a common description of the data, using standard terms and data structures. The recipient of the data should know exactly what the numbers represent and what is their history.
The discussion during the two workshops also coincided with the publication of a white paper on ethics by an organisation called Climate Service Partnerships (CPS). Many of the ideas from the IPCC workshop, the biodiversity meeting in Bogota, and points made in the CSP white paper all come together: better guidelines, best practices, and collaboration are necessary to avoid mal-adaptation to climate change.
I learned from the Bogota meeting that biologists often use a dataset called ‘worldclim‘ to provide a basis for climate information. Climate scientists then need to explain why worldclim often is not appropriate for describing local climatic conditions. The reason is that future projections are derived (interpolated) from coarse global climate models which do not account for local details such as geographical details, that many of the station records used to estimate the baseline may not have been quality checked, and that there are many regions with missing observations.
The concept of scales may also cause some confusion, with different definitions in different disciplines. The climate scientists need to know what exactly is the question and what kind of answers people expect. Also that there is a crucial difference between data and information, and that people often want an answer or some information rather than data. However, those who use climate data for further processing may have to adapt their analysis to the available information.
One example is a person who asks for hourly precipitation in order to figure out how often do we get a flash floods. So it is not really hourly data that is needed, but instead the answer to the question whether flash floods will become more frequent or severe. In other words, we may make sense out of rare events and extremes if we know how to pose a question that can be answered with science or statistics.
In other words, we have both information and knowledge that can be used as guidance in decision-making and climate change adaptation. However, we need to rethink our questions and look for cases where climate science can provide reliable information that have a direct relevance, even if we cannot get a complete answer. At least, we should look for ways to improve the information basis for decision making by looking at the type of information and data that has been used in the past. One way to do that is through a dialogue and co-production of knowledge.