New satellite observations to improve monitoring and forecasting of severe smoke pollution over SE Asia caused by Indonesian landscape burning

Lead Research Organisation: King's College London
Department Name: NCEO-King’s

Abstract

Outdoor air pollution is the world's greatest environmental health risk, and is contributed to greatly in the developing world by landscape burning. Smoke from landscape fires is a severe environmental health issue in parts of southeast Asia.. The smoke impacts are most keenly felt during dry spells in Indonesia, which hosts more than 80% of SE Asia's tropical peatlands. Over the last 50 years, deforestation and the construction of drainage canals, has increased the flammability of large areas of Indonesia's tropical peatland, during dry spells the peat can dry sufficiently to be ignitable by surface vegetation burning. The resulting fires can propagate downward into the peat and laterally, generating large areas of burning with fuel consumptions per unit area 100 times higher than other landscape fires. This releases copious amounts of particulate matter (PM) which adversely affects human health because the particles are small enough to pass into the human respiratory system causing respiratory and cardiovascular morbidity and mortality
This project teams researchers in the UK from STFC and KCL with universities ITB and BMKG to address this problem using satellite observations. In Indonesia BMKG has the remit within Indonesian to provide air quality information. However, this is based on a limited number of in situ monitoring stations across the nation, and no operational use of satellite aerosol data to complement and extend in situ coverage. Furthermore, air quality (AQ) forecast models need to be driven by accurate emissions magnitudes and timings. Unlike emissions from urban sources, those from peat burning cannot be prescribed from inventories, but must rather be based on real-time satellite observations since these fires are highly variable over space and time. Thus a reliable and timely source of fire and emissions data need to be provided, and if this is done then accurate forecasts of air pollution episodes should be possible using via an atmospheric chemical transport model (CTM) fed with these data. Such AQ forecasts alert authorities to take action (e.g. cancel outdoor activities, close schools, distribute personal protection masks).
Satellite-based approaches to estimate fuel consumption are available and mature. However, emissions factors have so far been based on laboratory combustion of just a few samples of tropical peat exhibiting significant variability. Furthermore, aerosol optical properties used in satellite retrievals of AOD (Aerosol Optical Depth) are based on measurements made in South American forests. Neither the emission factors nor the smoke optical properties will be fully representative of Indonesian 'wildfires', and especially not the extreme combustion conditions exist during El Nino periods - such as for example intense flaming forest fires atop large areas of burning peat. Only in situ sampling of smoke from representative wildfires can deliver emissions factors and aerosol optical properties representative of such conditions.
This project has five main objectives focused on achieving a step change in the quality of AQ measurements, monitoring and forecasts over Indonesia and importantly to develop the capacity for joint Indonesian and UK research in the area through joint field campaigns and visiting scientists. The first objectives focus on obtaining a unique set of ground measurements bespoke to Indonesian scrubland, forest, peat and agricultural residue fires. The subsequent objectives will capitalize on these new pieces of information to produce new satellite aerosol optical depth (and atmospheric particulate matter) information and fire radiative power-based fire emissions products capable of being used in Indonesia in real-time to support AQ assessments and model forecasts, and finally we will assess the quality of the full set of information produced - and in particular the atmospheric PM distributions.

Publications

10 25 50