cc: d.viner@uea date: Tue Nov 14 10:00:33 2000 from: Tim Osborn subject: MAFF response thresholds to: martin.parry@uea.ac.uk, s.park-dwyer@uea.ac.uk Sarah and Martin, David and I have finished reading the 60 pages of the response threshold reports. We have attempted to extract the climate changes that cause the economic response thresholds to be exceeded, for the five crop types used by Rounsevell and Jones. (1) Winter wheat. Harrison and Butterfield show responses to a range of variables (temperature, precipitation, solar radiation, vapour pressure, wind), but few of them change sufficiently in the climate change scenarios to even approach the threshold values. The exception is temperature, where winter warming on its own would be enough to lower yields and breach thresholds. *But*, winter precipitation and enhanced CO2 (Wheeler) offset the lower yields and prevent any thresholds from being reached. (2) Winter/spring barley. Harrison and Butterfield show that warming will enhance yields, but not enough to breach upper thresholds. Precipitation is unimportant. No lower thresholds will be breached either. (3) Grass ley. Warming will increase yield, so no lower thresholds are breached and Rounsevell and Jones do not define any upper thresholds. (4) Sugar beet. For reasonable levels of climate change, no cases reach even a 10% decrease in yield, so no thresholds are breached. (5) Lettuce. Wheeler's table 5 indicates yield reduction thresholds can be exceeded by temperature changes that are realisable. But when CO2 changes are also included (tables 6 & 7), yields are kept well above the thresholds. None will be breached. Do you have any comments on the above? Feel free to distribute this summary amongst the group. I will now concentrate on the last few force thresholds and give David all the information necessary to do the emission pathways work. To do this, I need to know what constitutes an exceedance of a threshold. Most of the graphs show the fraction of years for which a weather/climate event occurs. Was it discussed at the meeting how frequently the events could be tolerated? We could use a rather arbitrary value, such as 0.5, but in reality some events are worse than others so the threshold should perhaps be different. Also, a couple of cases are a little different. 4(iii), 4(iv,winter wheat) and 4(iv,spring wheat) are expressed as the average number of days per year (during the appropriate growth stage) that exceed 31C or 15mm. A value of 0.5 means an average of one day every two years, though this isn't the same as half the years being affected by an extreme since some years might have more than 1 event in them, leaving more than half with no events. I chose to express these daily extreme thresholds in this way because it seemed (at the time) the best way. The question then is: what frequency should the threshold be put at in each case? Regards Tim