Training on Trainers Workshop on Climate Change Analysis and Risk Communications

Workshop Objectives

The workshop was convened in 2008 at the Climate Systems Analysis Group in the University of Cape Town as part of the ACCCA project which is coordinated by UNITAR. The overall objective of the workshop is to provide a venue for hands-on work and guidance on the analysis and interpretation of climate data, and on understanding the regional contexts of climate change as a threat multiplier. The working and discussion sessions designed for these days focussed on:

• Developing a baseline package of a region's information.
• Understanding issues of meta-concepts involved in the interpretation of climate data, including confidence and uncertainty in climate change related to geographical and temporal scale, as well as the robustness of climate modeling information.
• Developing an understanding of the nature of scenario and projection production, including downscaling.
• Using the CCE tool to support an informed discussion and communication of climate change issues and vulnerability, impacts and adaptation options to decision-makers.

See also the Summary of Proceedings of the Training of Trainers Workshop on Climate Change Science 

Vocabulary and definitions from the workshop

This is the list of vocabulary which participants felt it would be useful to define. 

• Albedo The fraction of solar radiation reflected by a surface or object. Snow-covered surfaces have a high albedo, the albedo of soils ranges from high to low and vegetation-covered surfaces and oceans have a low albedo. The earth's planetary albedo varies mainly through varying cloudiness, snow, ice, leaf area and land cover changes. Albedo is an importantfeedback in the climate system. IPCC WGI glossary.

• Average can refer either to the mean, median or mode but mean is most commonly meant when the word average is used. The mean is calculated by adding all numbers in a dataset together and dividing by how many numbers there are. The average temperature in July for the period 1960-1990 means that all the July temperatures from that period have been added up and divided by the number of years to find a value which is representative of July temperature over the period.

• Anomalies An anomaly is the difference between the expected result and the actual result. The temperature anomaly in 2050 is the difference between the predicted temperature in 2050 and the average temperature in the present.

• Baroclinic conditions occur when the atmosphere is disturbed and there are temperature or pressure gradients. Baroclinic conditions are characterised by cold/warm fronts, troughs and an unstable atmosphere. Baroclinic areas of the atmosphere are generally found in the mid-latitudes. [1]

• Barotropic conditions are stable with little or no pressure or temperature gradient. They produce stable conditions generally occur in the tropics. [2]

• Boundary conditions Models cannot simulate all of reality, they must choose an area and model the processes within that area, so they have boundaries . Boundary conditions are how the model simulates the physical processes in this boundary region, and can have an impact on the way the rest of the simulation works. [3]

• Climate is the combination of all the variations in weather at a given location over a period of time. It is the average conditions that are expected at a location. The period of time for measuring climate is defined by the WMO as 30 years, with the average conditions from 1960-1990 often used to define 'current' climate. IPCC WGII glossary.

• Climate Sensitivity is the change in the climate system that would occur in response to a given external forcing. It often refers to what change would occur (or how sensitive the climate is) under different concentrations of carbon dioxide. IPCC WGI glossary.

• Dekad A ten day period.

• Diagnostic and Prognostic variables Prognostic variables are ones which are modelled using the internal physics of the model, such as temperature. Diagnostic models , such as rainfall, are ones which cannot be modelled dynamically, so are parameterized (see below). Projections of changes in prognostic variables are more reliable than those in diagnostic variables.

• Extreme events An extreme weather event is an event that is rare at a particular place and time of year. An extreme event would usually be rarer than the 10th or 90th percentile. Extreme events vary from place to plece; what is extreme in one area might be common in another. Single extreme events cannot be attributesd as being caused by climate change, as they may have occurred naturally, but climate change is expected to increase the occurence of extreme events. IPCC WGI glossary.

• Forecast A forecast is an expected outcome based on established patterns (physical, technological, economic, social etc) and is for the near-future - in climate terms forecasts can be made up to the level of seasons. IPCC WGIII glossary.

• GCM General Circulation Models (GCMs) are a class of computer-driven models for understanding climate and projecting climate change, where they are commonly called Global Climate Models. They are numerical representations of the real world which simulate actual processes, but do not capture all the complexity of the system. IPCC WGI glossary.

• GCM skill This is a measure of how good GCMs are at modelling particular processes.

• Gridded data the fundamental unit of gridded data is the cell, which represents a location in a continuous space set. The condition of a given cell is recorded as a numeric value for each cell. In spatial terms, gridded data are called rasters.

• Initial conditions are the state of the variables of the current climate which are used as the starting point for running GCMs. As we do not understand the whole of the climate system, these initial conditions are informed guesses and so can introduce uncertainty into GCM projections. To overcome this some groups run GCMs lots of different times, with slightly different initial conditions to give a range of outcomes.

• Interpolation is a method of constructing new data points (and surfaces) within the range of a discrete set of known data points. Two broad categories of interpolation methods area available for spatial data, differing in the level of detail gained, the weighting approach used and their treatment of observed values. These categories are discussed in detail below, but include: Deterministic methods, which use polynomial functions to fit a surface onto a set of points. These include the methods of Inverse Distance Weighting and Splines, and Geostatistical – which determine a priori trends in the data and then applies these trends to the task of surface fitting. Kriging techniques are an example of this method.

• ITCZ The Inter-Tropical Convergence Zone is a region of rising, unstable air near the equator which occurs when the NE and SE Trade Winds meet. The ITCZ migrates seasonally following the zone of greatest heating, and because it is an unstable mass of air causes significant rainfall. Changes in the way the ITCZ moves could have severe consequences on rainfall in many parts of Africa. [4]

• Lapse Rate The rate of change of an atmospheric variable, usually temperature, with height. A high lapse rate means change is rapid as temperature increases. [5]

• Monsoon A vital part of the climate system for many regions of the world, including S. Asia and W. Africa. A monsoon is the reversal in surface winds caused by different pressure gradients between the land and sea in summer and winter. In the dry phase of the monsoon circulation there is high pressure over the land surface and low pressure over the ocean, so winds blow from the land to the sea. In the wet phase, there is low pressure over the land (as hot air rises) and higher pressure over the sea, so the wind blows from the sea to the land carrying large amounts of moisture which are then deposited as rainfall over the land. [6]

• Normal Often used to describe the average conditions of a site. 'Normal climate' is often used to mean 'Average climate'.

• Parameterization This refers to the technique in climate models of representing processes that can't be resolved by the dynamic physics of the model, or that are below the scale at which the model operates, by static relationships between the large-scale processes and the process being modelled. For example a model may not be able to create rainfall because it occurs at a scale which is too small, so it might create a relationship which says that when humidity is above 70% it will rain. IPCC glossary.

• Prediction A prediction is an estimate of the way a system will react in the future, for example at a seasonal level. Predictions use the state of variables in the current system (for example sea-surface temperatures) to estimate the evolution of the system in the near-future. IPCC WGI glossary.

• Projection Looks at the response of the climate system to scenarios of changes in emissions, aerosols and radiative forcing. Projections depend on the assumptions underlying the scenario and model used, which may or may not occur, so are only possible futures as they are subject to uncertainty. IPCC glossary.

• Probability Density Function (PDF) This is a graphical representation of how often a certain outcome occurs, or is expected to occur.

• Probability is the likelihood, or chance that an event will occur. It is not yet possible to give climate projections probabilities because of the uncertainty present in the climate models.
• Radiative Forcing is the change in the net irradiance (in watts) at the tropopause due to an external driver such as an increase in greenhouse gasses. IPCC WGI glossary.

• Robustness The property of statistical procedures that are insensitive to small departures from the assumptions on which they depend, such as the assumption that certain distributions are normal. [7]

• Scenarios A scenario is a plausible description of how the future may develop, it is not a prediction of the what the future will be. They are based on a coherent set of assumptions about the driving forces and key relationships that determine what the future will look like. IPCC WGI glossary.

• Spatially Smoothed- a set of statistical methods used to fit spatial surfaces, such as those of a GCM to a set of observations (such as station data).

• Stationarity is an issue in empirical downscaling and refers to the assumption that relationships which exist between variables at present will stay the same into the future.

• Stochastic A stochastic event is based on random behavior. The occurrence of individual events cannot be predicted, although measuring the distribution of all observations usually follows a predictable pattern. These patterns can be described by statistical means. An example is the decay of radio active material, where a clump of matter has a measurable and thus predictable half-life time. It is impossible, however, to mark an individual atom and predict when it will decay and emit radiation. The latter process is a stochastic event. [8]

• Synoptic circulation is the large-scale circulation such as fronts and cyclones which drive the climate system.

• Surprise Strictly speaking, a surprise is an outcome cannot be anticipated; by definition it is an unexpected event. However, what gets labeled as `surprise' depends on the extent to which what happens departs from community expectations and on the salience of the problem [9]

• Uncertainty This is the unpredictability of a system or model. Uncertainty in models can come from many sources; a lack of information, lack of technical ability to model complex processes, a lack of knowledge about a system or uncertainty over human behaviour.

• Weather is the actual state of the atmosphere at a given time in a given place. It is what people experience on a day to day basis.

• Weather Generator Based on the statistical characteristics of observed weather at a site, a weather generator can produce long-term forecasts of weather at that site. [10]

• Teleconnections A connection between climate variations over widely separated parts of the world, so that climate variations in one part of the world will cause the climate system in a different part of the world to respond in a certain way.

• Thresholds A threshold is a point in a system at which external forcing of system (for example due to increasing greenhouse gasses) causes a reorganisation of the system into a significantly different state from which it is impossible, or takes a very long time, to return to the original state. For example warming may switch ocean circulation in the N. Atlantic into a different state in which the thermohaline circulation does not transport as much heat to N. Europe, and it would take several hundred years to switch back. IPCC WGII glossary