Editing IPCC:AR6/WGI/Chapter-10 (section)

==== 10.2.1.2 Derived Products ====

<div id="h3-7-siblings" class="h3-siblings"></div>

Derived observational products are created from raw datasets collected from surface stations, remote-sensing instruments, or research vessels, which are converted into meaningful physical quantities by applying a suitable measurement theory, using either statistical interpolation techniques ( [[#10.2.2.4|Section 10.2.2.4]] ) or numerical atmospheric and land surface models ( [[#Bosilovich--2015|Bosilovich et al., 2015]] ).

Most global observational datasets are available at coarse temporal and spatial resolution, and do not include all available station data from a particular region, due to data availability problems. Therefore, efforts have been made to develop regional or country-scale datasets (Annex I). Radar and satellite remote sensing are resources that can provide a valuable complement to direct measurements at regional scale. Examples for precipitation have been described already, some of which have been released to the community ( [[#Dinku--2014|Dinku et al., 2014]] ; [[#Oyler--2015|Oyler et al., 2015]] ; [[#Manz--2016|Manz et al., 2016]] ; [[#Dietzsch--2017|Dietzsch et al., 2017]] ; [[#Yang--2017|Yang et al., 2017]] ; [[#Bližňák--2018|Bližňák et al., 2018]] ; [[#Krähenmann--2018|Krähenmann et al., 2018]] ; [[#Panziera--2018|Panziera et al., 2018]] ; [[#Shen--2018|Shen et al., 2018]] ). However, some of these datasets are limited by their short record, varying between one ( [[#Shen--2018|Shen et al., 2018]] ) and 64 years ( [[#Oyler--2015|Oyler et al., 2015]] ).

Reanalysis products are numerical climate simulations that use data assimilation to incorporate as many irregular observations as possible. These products encompass many physical and dynamical processes. They generate a coherent estimate of the state of the climate system on uniform grids either at global ( [[#Chaudhuri--2013|Chaudhuri et al., 2013]] ; [[#Balsamo--2015|Balsamo et al., 2015]] ), regional ( [[#Chaney--2014|Chaney et al., 2014]] ; [[#Maidment--2014|Maidment et al., 2014]] ; [[#Dahlgren--2016|Dahlgren et al., 2016]] ; [[#Langodan--2017|Langodan et al., 2017]] ; [[#Attada--2018|Attada et al., 2018]] ; [[#Mahmood--2018|Mahmood et al., 2018]] ) or country scales ( [[#Rostkier-Edelstein--2014|Rostkier-Edelstein et al., 2014]] ; Krähenmannet al., 2018; [[#Mahmood--2018|Mahmood et al., 2018]] ).

Reanalyses incorporate an increasing volume of observations from a growing number of sources over time, which sometimes presents a difficulty for trend analysis. However, regional reanalyses are valuable for regional climate assessments, since they can employ high-resolution model simulations due to their limited spatial domain. Their accuracy is also better than global reanalyses since they are often developed over regions with a high density of observational data (sometimes not freely available for all regions) to be assimilated into the model (e.g., [[#Yamada--2012|Yamada et al., 2012]] ). Regional reanalyses can assimilate locally dense and high-frequency observations, such as from local observation networks ( [[#Mahmood--2018|Mahmood et al., 2018]] ; [[#Su--2019|Su et al., 2019]] ) and radar precipitation ( [[#Wahl--2017|Wahl et al., 2017]] ) in addition to the observations assimilated by global reanalyses. In some regional reanalyses, satellite-derived high-resolution sea ice ( [[#Bromwich--2016|Bromwich et al., 2016]] , 2018) and sea surface temperature ( [[#Su--2019|Su et al., 2019]] ) are also applied as lower boundary conditions. The periods of regional reanalyses are limited by the availability of the observations for assimilation and by the global reanalyses needed as lateral boundary conditions. Most regional reanalyses cover the past 10 to 30 years. There are also regional reanalysis activities that use conventional observations only, which produce consistent datasets over 60 years to capture precipitation trends, extremes and changes ( [[#Fukui--2018|Fukui et al., 2018]] ). Existing regional reanalyses cover North America ( [[#Mesinger--2006|Mesinger et al., 2006]] ), Europe ( [[#Dahlgren--2016|Dahlgren et al., 2016]] ; [[#Jermey--2016|Jermey and Renshaw, 2016]] ; [[#Kaspar--2020|Kaspar et al., 2020]] ), the Arctic ( [[#Bromwich--2016|Bromwich et al., 2016]] , 2018), South Asia ( [[#Mahmood--2018|Mahmood et al., 2018]] ), and Australia ( [[#Su--2019|Su et al., 2019]] ). A project for regional reanalysis covering Japan has also started ( [[#Fukui--2018|Fukui et al., 2018]] ), where grid spacing is between 5 and 32 km, although cumulus parametrizations are still needed to compute sub-grid scale cumulus convection. Recently, reanalyses using convection-permitting regional models have been published (e.g., [[#Wahl--2017|Wahl et al., 2017]] , for central Europe).

The data assimilation schemes used in regional reanalyses are often relatively simple methods, specifically nudging ( [[#Kaspar--2020|Kaspar et al., 2020]] ) and 3DVAR ( [[#Mesinger--2006|Mesinger et al., 2006]] ; [[#Bromwich--2016|Bromwich et al., 2016]] ; [[#Dahlgren--2016|Dahlgren et al., 2016]] ), rather than the more complex schemes implemented in state-of-the-art global reanalysis systems. This is partly due to limitations of computational resources. Recently, a number of regional reanalyses using more sophisticated methods, such as 4DVAR and Ensemble Kalman filter, have been published ( [[#Jermey--2016|Jermey and Renshaw, 2016]] ; [[#Fukui--2018|Fukui et al., 2018]] ; [[#Mahmood--2018|Mahmood et al., 2018]] ; [[#Su--2019|Su et al., 2019]] ). The regional reanalyses also incorporate uncertainties due to deficiencies of the models, data assimilation schemes and observations. To estimate uncertainties, some regional reanalyses apply data assimilation using ensemble forecasts ( [[#Bach--2016|Bach et al., 2016]] ). Another approach compares multiple regional reanalyses produced with different systems covering the same domain, which represents the uncertainties better than single reanalysis systems with ensemble data assimilation schemes ( [[#Kaiser-Weiss--2019|Kaiser-Weiss et al., 2019]] ).

The regional reanalyses represent the frequencies of extremes and the distributions of precipitation, surface air temperature, and surface wind better than global reanalyses ( ''high confidence'' ). This is due to the use of high-resolution regional climate models (RCMs), as indicated by different regional climate modelling studies ( [[#Mesinger--2006|Mesinger et al., 2006]] ; [[#Bollmeyer--2015|Bollmeyer et al., 2015]] ; [[#Bromwich--2016|Bromwich et al., 2016]] , 2018; [[#Dahlgren--2016|Dahlgren et al., 2016]] ; [[#Jermey--2016|Jermey and Renshaw, 2016]] ; [[#Fukui--2018|Fukui et al., 2018]] ; [[#Su--2019|Su et al., 2019]] ). Regional reanalyses, however, retain uncertainties due to deficiencies in the physical parametrization used in RCMs and by the use of relatively simple data assimilation algorithms ( [[#Bromwich--2016|Bromwich et al., 2016]] ; [[#Jermey--2016|Jermey and Renshaw, 2016]] ; [[#Su--2019|Su et al., 2019]] ). Regional reanalyses can provide estimates that are more consistent with observations than dynamical downscaling approaches, due to the assimilation of additional local observations ( ''high confidence'' ) ( [[#Bollmeyer--2015|Bollmeyer et al., 2015]] ; [[#Fukui--2018|Fukui et al., 2018]] ).

<div id="10.2.2" class="h2-container"></div>

<span id="challenges-for-regional-climate-change-assessment"></span>