Insights into ECOSSE modelling of soil organic carbon at site scale from Irish grassland sites and a French grazed experimental plot

ABSTRACT

The work provides insights into soil organic carbon (SOC) modelling procedures associated with different management practices for Irish grassland sites selected from two large soil databases (LUCAS-2009 _[1] and Teagasc-SIS _[2]) and a single treatment-plot  from France  (paddock of a long- term grassland-experiment) _[3]. Modelling of SOC was done at site scale using “Model to Estimate Carbon in Organic Soils -Sequestration and Emissions” (ECOSSE) 6.2b version of the model in site-specific mode _[4]. The selection of Irish sites and the Irish model input-parameters followed procedures explained in Premrov et al. (2020) _[5]. As explained in Premrov et al. (2020) _[5], special attention was given to model SOC-input data because the preliminary findings showed high sensitivity of model predictions to the initial SOC-inputs _[5]. Initial SOC-inputs for Irish sites were extracted from the Irish soil NSDB-database _[8] because of lack of data at that time. The preliminary SOC modelling results from Irish sites [RMSE >36%; 84 sites (out of total 95 pre-selected LUCAS and SIS sites after excluding 11 potential outliers _[7])]indicated that further work is needed on obtaining initial SOC-input data. The new LUCAS-2015 _[6] soil-point data in combination with older LUCAS-2009 _[1] data provide opportunities to resolve this issue, which is currently work in progress. Considering that Irish sites were selected from large soil-databases that lacked detailed site-specific information (i.e. stocking rates and fertilisation-data could be obtained only in general form), the treatment-plot from France _[3] was also simulated to gain further insights into the ECOSSE SOC modelling at site/point-scale. This work confirmed the importance of using appropriate conversion-factor when applying stocking rates as a proxy for the manure-inputs (as an alternative for grazing _[7]). Further insights included the importance of assessing the modelled SOC ‘trends’ over time, and its comparison with observed ones _[7].

 

Acknowledgements

SOLUM project is funded under the Irish EPA Research programme 2014-2020. Thanks go to Dr Jo Smith (University of Aberdeen, Scotland) for ECOSSE-model and to all who provided data or advice/support, among others, Teagasc-SIS, Ireland;  French national-observatory SOERE ACBB, a part of ANAEE-F French national infrastructure, and Dr Katja Klumpp (INRAe, France).

 

Literature

 

[1] JRC (2018). LUCAS-2009, ESDAC. JRC. EC.

[2] Teagasc, (2018) Irish Soil Information System (SIS). Teagasc, EPA, Ireland.

[3] SOERE ACBB (2020) http://www.soere-acbb.com/

[4] Smith, J., et al. (2010). ECOSSE. User Manual.

[5] Premrov, A., et al. (2020). Insights into modelling of soil organic carbon from Irish grassland sites using ECOSSE model. EGU2020-8090. doi.org/10.5194/egusphere-egu2020-18940; (CC-BY-4).

[6] JRC (2020). LUCAS 2015, ESDAC. JRC. EC.

[7] Saunders, M. et al. (draft-report 2020) SOLUM. EPA Research Report. 2016-CCRP-MS.40.

[8] EPA (2007). National Soils Database (NSDB). Environmental Protection Agency, Ireland.