Workpackage number			4		Start date or starting event: full duration of project
Workpackage title: Fire-Climate-Carbon cycle interactions on regional and continental scale
Activity Type: RTD/Innovation activities
Participant No.	1	2		3		4	5	6	7	8	9	10	11	12	13	14	15	16
Person-months:	6	0		30		2	0	0	5	0	0	7	0	0	0	0	19	24

Objectives The overall objective of the Fire WP is to quantify the current African fire-related GHG emissions and fire-vegetation feedbacks, and their regional and interannual variation using an integrated remote sensing and process-based modelling methodology.   4.1 Development of an improved and validated method to estimate C emissions from forest and savanna fires, combining the three discrete, multi-sensor EO-based approaches of active fire detection, burned area mapping, and fire radiative energy assessment.   4.2 Derivation of key datasets of fire variability and an analysis of influencing factors, including climatic oscillations (e.g.. El Nino), population density, land cover type and meteorology.   4.3 Analysis of driving factors of the fire regime in different parts of the study region to improve fire modelling through quantification of coupling of interannual climate variability, population density, land cover type and fire frequency and intensity using a remote sensing data analysis approach.   4.4 Incorporation of processes identified in (4.1) - (4.3) into an improved fire module of a DGVM model framework (LPJ-GUESS). Regional and continental model estimates of total fire carbon loss, post-fire carbon-dynamics and fire-climate-vegetation interactions.

Description of work Task 4.1 Interactive climate-fire-vegetation effects.   Task 4.2 Remote sensing of fires and estimates of carbon losses.

Deliverables 4.1 A first full year (2004) of fire radiative power and energy data for the whole of African continent, at source (~3 km) spatial resolution and 15 minute temporal resolution, derived from SEVIRI and compared to MODIS in the study regions; together with a sensitivity analysis of the FRP retrievals to parameterisation errors and algorithmic assumptions (Month 13).   4.2 Study region, MODIS-derived FRP datasets for 2002 and 2003 (Month 18).   4.3 Burned area map of study region from remote sensing (Month 18).   4.4 Integrated methodology combining burned area, active fire detection and FRE into a single, optimised biomass burning C emissions product for Africa (Month 22-27).   4.5 Field-tested relationship between fire radiative energy and carbon emission for different African fuel types and a range of fuel moistures, enabling the above FRP datasets to be converted to carbon flux estimates from identified fires. Validate the fire module of the LPJ-GUESS model using new remote sensing and GIS data, as well as information on population density (Month 22-27).   4.6 Process-understanding of regional and continental fire C losses in Africa, their relation to climate, area burnt and frequency and causes for interannual variation. Asessment of fire-climate-vegetation interactions on boundaries, e.g., savanna-forest and savanna-desert that can be considered particularly vulnerable to climate change and human pressure (Month 22-30).

Milestones and expected result 4.1 WP kick-off workshop in Skukuza, relatively early on in the project to discuss our various approaches, including a one or two-day course or so at Wits afterwards (Month 4).   4.2 Algorithms for fire identification and characterisation; Improved fire routine in LPJ-GUESS; Revised PFT in LPJ GUESS, accommodating fire characteristics (Month 12).   4.3 Analysis of matching SEVIRI- and MODIS-derived FRP/FRE datasets for 2004. Burned area map of study region over 3 years (e.g., 2002-2004) and matching MODIS-derived FRP data; Analysis fire-human interactions (e.g., population density, land use, use of controlled fire) and improved assessment of human impacts on fire in LPJ-GUESS (Month 18).   4.4 Validation of fire products using ground and airborne data, and comparison to model results; First continental LPJ-GUESS model runs; 2nd WP workshop in West Africa, again with training workshop, lectures (Month 24).   4.5 Estimation of carbon emission from fires, and its spatial and temporal variation in study regions/years (Month 27).   4.6 Analysis of causal factors in the fire regime using GIS and climate data, comparison with model causal analysis. Regional and continental model runs to estimate the effects of fire on African carbon balance (Month 30).   4.7 Completion of manuscripts, publications (Month 36).