Claire Granier LATMOS/IPSL, Paris, France Currently on-leave at the Max Planck Insitute for Meteorology, Hamburg Also at the NOAA Earth System Research Laboratory, Boulder, USA Emission Inventories for Research Projects: Focus on Global Anthropogenic Emissions

Different types of emissions for different projects: Forecast of the atmospheric composition, observations from campaigns  Wide range of chemical species  high spatial and temporal resolution Global scale, long-range transport  limited number of chemical species  moderate spatial and temporal resolution  long-term variation (a few decades)  need some coupling emissions/meteorological conditions Climate studies: impact of climate on emissions and of emissions on climate  long-lived species, aerosols and a few ozone precursors  emissions models or algorithms to take into account land-use changes and human-related changes  past/future realistic scenarios (decades-century)

List of most publicly available global inventories Publicly available regional inventories exist for Europe, USA, Canada, India and China

The ACCMIP/MACCity dataset used in MACC Linked to the development of the emissions used in the IPCC 5 th Assessment report (2013)  Be as consistent as possible with other developments/projects More details: - Lamarque et al., ACP van Vuuren et al., Climatic Change, Granier et al., Climatic Change, Lamarque et al, Climatic Change, 2011

ACCMIP = Consistent long-term emissions dataset [emissions for Atmospheric Chemistry and Climate Modeling Intercomparison Project] Developed in by an international group ACCMIP emissions dataset:  Developed in support of the IPCC AR5 report  Anthropogenic and biomass burning emissions  Period covered: 1850 – 2000  Avalaible for each decade  no seasonal variation (except for ships)  0.5x0.5 degree

MACCity (Emission dataset as part of MACC and CityZen) After 2000, no global emissions dataset available  Use of one of the IPCC future scenarios (RCP 8.5) for 2005, 2010 and Note that, since mid-August 2010, EDGAR v4 provides 2005 global emissions MACCity anthropogenic emissions:  Linear interpolation for of ACCMIP and RCP8.5 emissions  Implementation of seasonal emissions from the RETRO emissions (Schultz et al., 2007)

Sectors in the ACCMIP and MACCity datasets Different from the SNAP sectors in regional emissions

Are any of these emissions accurate?  A systematic evaluation of surface emissions has started within GEIA (Global Emissions Inventory Activity): geiacenter.org  Open to all people willing to propose new data / analyze results  Start with the period  Only publicly available gridded inventories considered so far  Focus on: CO, NOx, SO2 and BC (recently added: CH4, OC, total NMVOCs and NH3)  Assess whether one of the scenarios is closer than the others to regional emissions available after 2000, assuming regional emissions are more accurate

Examples of comparison of anthropogenic emissions

CO Western and Central Europe: Global inventories + EMEP and TNO EU inventories

CO USA: Global inventories + EPA US inventory CO China: Global and regional inventories

But, are regional emissions really more accurate  Based on reported data by countries/states/… CO traffic emissions from reported data CO emissions inferred from observations: systematic aircraft campaigns From Parrish, Atmos. Env More evaluations needed: - Inverse modeling, but limited to just a few species - Long-term monitoring/specific campaigns

SO2 USA: Global inventories + EPA US inventory SO2 China: Global inventories + REAS and ACCESS regional inventories

Biomass burning in Africa: CO and BC emissions AMMABB values:  Very different from other datasets Maybe the most consistent with: - Observations - Inverse modeling studies

Summmary: ratio between highest and lowest emissions Green: 1 to 1.25 Yellow: 1.25 to 1.5 Orange: 1.5 to 1.75 Grey: Above 1.75 BUT: consensus ≠ accuracy

CO emissions from different available global datasets for 2005: including EDGAR-v4 data  More work need for understanding post-2000 global emissions

A few examples of uncertainties in emission inventories - More details in: Granier, C., B. Bessagnet, T. Bond, A. D’Angiola, H. Denier van der Gon, G. J. Frost, A. Heil, J. W. Kaiser, S. Kinne, Z. Klimont, S. Kloster, J.-F. Lamarque, C. Liousse, T. Masui, F. Meleux, A. Mieville, T. Ohara, J.-C. Raut, K. Riahi, M. G. Schultz, S. J. Smith, A. Thompson, J. van Aardenne, G. R. van der Werf, D. P. van Vuuren (2011) Evolution of anthropogenic and biomass burning emissions of air pollutants at global and regional scales during the period, Climatic Change, DOI /s  Large diversity in space and time, difficult to quantify  Lack of information: non reported, not existing  Extrapolation errors when information is lacking (after 2000)  Most inventories are non-transparent about method and data  Difficulties to compare data for sectors of different definitions  Lack of different independent inventories  Lack of measurement data and model studies to confront inventories with

Where to find the emissions used in the intercomparison? Most data are available from the ECCAD database = Emissions of atmospheric Compounds: Compilation of Ancillary Data ECCAD= emissions database used in MACC and MACC-II

CU ETHER – 23/24 Novembre 2009 A few maps of ancillary data

CU ETHER – 23/24 Novembre 2009 A few examples of emission maps

ECCAD – Emissions Totals Global Total NOx, Residential, year 2005 : 6.24 Tg/year Totals for different regions Total by regions : from 0.1 to 1.18 Tg/year Output in excel/csv

A few issues to be looked at in MACC-II (if some data available) VOCs speciation  Only total VOCs reported in inventories  Atmospheric models need to know the detailed speciation on emissions  Models use their own speciation, but how accurate is it?

Ratios Comparison Ratios Comparison: LONDON Emissions and Concentrations Traffic site Mixed site

How to implement such temporal profiles? detailed data only easily and freely available for UK Are there any recent data available? Season VOCs CO weekly emissions Temporal variation NOx diurnal emissions

PM Marine aerosols (NaCl) Biological material (pollen, spores, …) H2OH2O Resuspended dust (crustal + tires, breaks,...) Inorganic emissions - minerals (SiO 2, …) - metals (Pb, Fe, Zn, …) Elemental C, particulates Organic C, particulates Semi-volatile organic compounds Organic C, gaseousNH 3 SO 2,SO 4 -- H 2 SO 4 NO x NH 4 NO 3 (NH 4 ) 2 SO 4 How to speciate PM emissions ?

Ship emissions: large discrepancies between global and regional emissions  Will be addressed in MACC-II in collaboration with the ACCESS European project (management of the Arctic) Changes in NOx emissions in Europe : ships will be the largest contributor to NOx emissions by ~

Examples of Coordinated Emissions Activities: colaborations EU, USA and Canada Will be discussed at the IWAQFR workshop in Potomac, USA November 2011

Research Progress Issues that Limit Emissions Research Where We Are Where We Want to Go Documented Data, Data Access, & User- Friendliness Species- and Region- Dependent Inventory Methods Grid Inventories vs. Sector and Area Totals Infrequent Temporal Availability Emission Trends vs. Method Changes State, National, Regional and Global Data Not Compatible

Additional slides

Ratios Comparison In the Developed world… Los Angeles, CA, USA Tokyo, Japan New York, NY, USA

CO emission inventory VOC emission inventory CO & VOC observational data VOC/CO ratios Developed countries Developing countries VOC/CO ratios How to improve the EI? Latin America Asia Africa North America Europe

Temporal variation of fossil fuel emissions Seasonality, diurnal, weekly Cultural differences too From RIVM, the Netherlands (in 2005) Seasonal diurnal

Kollamthodi_AEA_2008 Issue 5: emissions from ships

Evolution of transport-related emissions of particulate matter in the future Changes in NOx emissions in Europe

Ship emissions: large discrepancies between global and regional emissions

-Very large differences between global/regional models - Difficult to use these results to assess emissions - Global models not worse than regional models From Colette et al., ACPD, 2011 Can we use model results to assess emissions: Ozone trends from 4 regional and 2 global models

Mace Head (Ireland)