Editing IPCC:AR6/SRCCL/Chapter-2 (section)

== B2.2 Methodologies for estimating national to global scale anthropogenic land carbon fluxes ==

<div id="section-2-3-3-2-land-use-effects-block-1"></div>

'''Bookkeeping/accounting models''' calculate changes in biomass and soils that result from changes in land activity using data on biomass density and rates of growth/decomposition, typically from ground-based inventory data collection (field measurements of carbon in trees and soils) (Houghton et al. 2012 <sup>[[#fn:r709|709]]</sup> ; Hansis et al. 2015 <sup>[[#fn:r710|710]]</sup> ; Houghton and Nassikas 2017 <sup>[[#fn:r711|711]]</sup> ). The approach includes only those changes directly caused by major categories of land-use change and management. The models do not explicitly include the indirect effects to changing environmental conditions, although some effects are implicit in the biomass, growth rates and decay rates used. Thus, the models may overestimate past fluxes. The bookkeeping models include fluxes from peatland burning based on GFED estimates (Randerson et al. 2015 <sup>[[#fn:r712|712]]</sup> ).

'''DGVMs''' simulate ecological processes, such as photosynthesis, respiration, allocation, growth, decomposition etc., driven by environmental conditions (climate variability, climate change, CO <sub>2</sub> , nitrogen concentrations). Models vary with respect to the processes included, with many since AR5 now including forest management, fire, nitrogen and other management (Sitch et al. 2005 <sup>[[#fn:r713|713]]</sup> ; Le Quéré et al. 2018 <sup>[[#fn:r714|714]]</sup> ). Models are forced with increasing atmospheric CO <sub>2</sub> and changing climate, and run with and without ‘land use change’ (land cover and forest harvest) to differentiate the anthropogenic effects from the indirect effects of climate and CO <sub>2</sub> : the ‘land sink’. Thus, indirect effects are explicitly included. This approach also includes a ‘lost atmospheric sink capacity’, or the carbon uptake due to environmental effects on forests that does not happen once the forests are removed (Pongratz et al. 2010 <sup>[[#fn:r715|715]]</sup> ).

'''Integrated assessment models (IAMs)''' use storylines to construct alternative future scenarios of GHG emissions and atmospheric concentrations within a global socio-economic framework, including projections of AFOLU based on assumptions of, for example, crop yields, population growth and bioenergy use (Cross-Chapter Box 1 and Chapter 1). Some models include simplified DGVMs, which may include climate and CO <sub>2</sub> effects, while others use AFOLU emissions from other sources.

'''ESMs''' couple DGVMs, surface hydrology, and energy exchange models with atmosphere, ocean, and sea ice models, enabling exploration of feedbacks between climate change and the carbon cycle (e.g., warming effects increase soil and plant respiration and lead to higher atsmpheric CO <sub>2</sub> concentrations, which in turn promote plant growth) (Friedlingstein et al. 2014 <sup>[[#fn:r716|716]]</sup> ). They sometimes include numerical experiments with and without land-use change to diagnose the anthropogenic AFOLU flux (Lawrence et al. 2016 <sup>[[#fn:r717|717]]</sup> ).

'''Satellite data''' can be used as a proxy for plant activity (e.g., greenness) and to map land cover, vegetation fires and biomass density. Algorithms, models and independent data are used to calculate fluxes of CO <sub>2</sub> from satellite data, although calculating the net carbon flux is difficult because of the lack of information on the respiratory flux. Some active satellite sensors (LiDAR) are able to measure three-dimensional structure in woody vegetation, which is closely related to biomass density (Zarin et al. 2016 <sup>[[#fn:r718|718]]</sup> ; Baccini et al. 2012 <sup>[[#fn:r719|719]]</sup> ; Saatchi et al. 2011 <sup>[[#fn:r720|720]]</sup> ). Together with land-cover change data, these estimates of biomass density can be used to provide observational-based estimates of fluxes due to changes in forest area (e.g., Tyukavina et al. (2015) <sup>[[#fn:r721|721]]</sup> , Harris et al. (2015) <sup>[[#fn:r722|722]]</sup> and Baccini et al. (2012) <sup>[[#fn:r723|723]]</sup> or degradation (Baccini et al. 2017 <sup>[[#fn:r724|724]]</sup> )). Satellite estimates of biomass vary considerably (Mitchard et al. 2013 <sup>[[#fn:r725|725]]</sup> ; Saatchi et al. 2015 <sup>[[#fn:r726|726]]</sup> ; Avitabile et al. 2016 <sup>[[#fn:r727|727]]</sup> ): data are available only for recent decades, methods generally assume that all losses of carbon are immediately released to the atmosphere and changes in soil carbon are generally ignored. The approach implicitly includes indirect and natural disturbance effects as well as direct anthropogenic effects.

'''Atmospheric inversions''' use observations of atmospheric concentrations with a model of atmospheric transport, based on data for wind speed and direction, to calculate implied emissions (Gatti et al. 2014 <sup>[[#fn:r728|728]]</sup> ; Liu et al. 2017a <sup>[[#fn:r729|729]]</sup> ; van der Laan-Luijkx et al. 2017 <sup>[[#fn:r730|730]]</sup> ). Since AR5, there has been an increase in availability of concentration data from flux tower networks and satellites, enabling better global coverage at finer spatial scales and some national estimates (e.g., in the UK inverse techniques are used together with national GHG inventories). A combination of concentrations of different gases and isotopes enables the separation of fossil, ocean and land fluxes. However, inversions give only the net flux of CO <sub>2</sub> from land; they cannot separate natural and anthropogenic fluxes.

'''Micrometeorological flux measurements''' data on CO <sub>2</sub> concentrations and air movements recorded on instrumented towers enable the calculation of CO <sub>2</sub> flux at the ecosystem scale. Global and regional Flux Networks (FluxNet (global), AsiaFlux, Ameriflux (North America), ICOS (EU), NEON (USA), and others) contribute to a global flux database, which is used to verify the results of modelling, inventory and remote sensing studies.

'''FAOSTAT''' has produced country level estimates of GHG emissions (Tubiello et al. 2013 <sup>[[#fn:r731|731]]</sup> ) from agriculture (1961–2016) and land use (1990–2016) using a globally consistent methodological approach based largely on IPCC Tier 1 methods of the 2006 IPCC Guidelines (FAO 2015 <sup>[[#fn:r732|732]]</sup> ). FAO emissions estimates were used as one of the three database inputs into the AR5 WGIII AFOLU chapter. Non-CO <sub>2</sub> emissions from agriculture are estimated directly from national statistics of activity data reported by countries to FAO. CO <sub>2</sub> emissions from land use and land-use change are computed mostly at Tier 1, albeit at fine geospatial scales to capture effects from peatland degradation and biomass fires (Rossi et al. 2016 <sup>[[#fn:r733|733]]</sup> ). Emissions from forest land and deforestation are based on the IPCC carbon stock change method, thus constituting a Tier 3 estimate relying on country statistics of carbon stocks and forest area collected through the FAO FRA. The carbon flux is estimated assuming instantaneous emissions in the year of forest area loss and changes in carbon stocks within extant forests, but does not distinguish ‘managed’ and ‘unmanaged’ forest areas, albeit it treats separately emissions from primary, secondary and planted forest (Federici et al. 2015 <sup>[[#fn:r734|734]]</sup> ).

'''Country Reporting of GHG Inventories (GHGIs)''' : All parties to the UNFCCC are required to report national GHGIs of anthropogenic emissions and removals. Reporting requirements are differentiated between developed and developing countries. Because of the difficulty of separating direct anthropogenic fluxes from indirect or natural fluxes, the IPCC (2003) <sup>[[#fn:r735|735]]</sup> adopted the ‘managed land’ concept as a proxy to facilitate GHGI reporting. All GHG fluxes on ‘managed land’ are defined as anthropogenic, with each country applying their own definition of ‘managed land’ (i.e., ‘where human interventions and practices have been applied to perform production, ecological or social functions’ (IPCC 2006) <sup>[[#fn:r736|736]]</sup> ). Fluxes may be determined on the basis of changes in carbon stocks (e.g., from forest inventories) or by activity data (e.g., area of land cover change management activity multiplied by emission factors or with modelled fluxes). Depending on the specific methods used, GHGIs include all direct anthropogenic effects and may include the indirect anthropogenic effects of environmental change (generally sinks) and natural effects (Section 2.3.1.2). GHG fluxes from ‘unmanaged land’ are not reported in GHGIs because they are assumed to be non-anthropogenic. The reported estimates may then be filtered through agreed ‘accounting rules’ (i.e., what countries actually count towards their mitigation targets (Cowie et al. 2007 <sup>[[#fn:r737|737]]</sup> ; Lee and Sanz 2017 <sup>[[#fn:r738|738]]</sup> ). The accounting aims to better quantify the additional mitigation actions by, for example, factoring out the impact of natural disturbances and forest age-related dynamics (Canadell et al. 2007 <sup>[[#fn:r739|739]]</sup> ; Grassi et al. 2018 <sup>[[#fn:r740|740]]</sup> ).

<div id="section-2-3-3-2-land-use-effects-block-3" class="box"></div>

<span id="b-2.3-co-2-fertilisation-and-enhanced-terrestrial-uptake-of-carbon"></span>