Understanding the phenomenon of clay shrinking and swelling by means of an indicator aggregated by commune: the magnitude of droughts
(1) Sophie Barthelemy - CNRM, Université de Toulouse, Météo-France, CNRS, Toulouse, France
(2) Bertrand Bonan - CNRM, Université de Toulouse, Météo-France, CNRS, Toulouse, France
(3) Gilles Grandjean - Bureau de recherches géologiques et minières
(4) David Moncoulon - Modelling, R&D Department, CCR
(5) Jean-Christophe Calvet - CNRM, Université de Toulouse, Météo-France, CNRS, Toulouse, France
INTRODUCTION
The phenomenon of shrinking and swelling of clays is the second most costly peril for the natural catastrophe compensation scheme, with more than 15 billion euros in compensation since 1989 [1]. As already illustrated by the 2022 drought in France, an increase in these costs is expected as a result of climate change, due to the amplification of annual soil drought-rehydration cycles [2].
It is in this context that a doctoral thesis was initiated in 2021, co-financed by Météo-France, BRGM and CCR. The goal is to develop new tools to estimate the loss experience related to the phenomenon of geotechnical dryness by drawing on the resources of the various partners. Over three years, the surface models developed by Météo-France, the geotechnical models and data from BRGM, and a national claims database compiled by CCR will be used.
In the first part of the thesis, the link between drought and claims is reviewed at commune level. One indicator, the magnitude, is constructed from soil wetness records at several depths produced by the ISBA[3,4] model developed by Météo-France. Soil moisture is measured at a spatial resolution of 8 km. The aim is to extract annual drought information, which is directly comparable via a rank correlation calculation[5] to aggregate loss data per year and per commune. This data is provided by CCR. The magnitude is obtained by calculating the annual integral of the soil wetness index (SWI) [6] under a certain threshold value associated with a drought frequency. The lack of knowledge on the mineralogy and water content of soils at plot level remains a scientific barrier to date, so a statistical approach is adopted. The magnitude is calculated for different SWI threshold values, corresponding to the percentiles of the statistical distribution of daily SWI. To vary the situations, different versions of the ISBA model are used, and different depths of the soil layer are considered. The results of the rank correlation calculation reveal the optimal configuration of the magnitude calculation.
The method was applied separately to five sets of four communes, located in different climates and geological contexts. The first results show that the best correlation with claims experience is obtained when considering the wetness of the soil layer between 1.5m and 2.0m depth, using low SWI threshold values. Shallower soil layers have a higher temporal variability, which is more related to quick weather variations.
Once the optimal configuration has been established, a classification[7] of the magnitudes provides estimates of the loss ratio. At this stage, the method does not allow for a more detailed characterisation of losses according to the type of building structures or the age of the buildings. The impact of the shrink-swell exposure class will be assessed in a next step. The magnitude coupled with future climate scenarios will allow the analysis of the impact of climate change on the loss experience of geotechnical droughts. These insurance projections will also be used to support the Nat Cat scheme./
# drought
# Clay shrinking-swelling (CSS)
# ISBA land surface model
# claims experience
Figure 1 - Method to compute the magnitude drought indicator from the ISBA model, and determine its optimal configuration
THE PARTNERS
The thesis is co-supervised by CNRM, BRGM and CCR. The CNRM (Centre national de recherches météorologiques) is a joint research unit (UMR 3589) formed by Météo-France and the CNRS. The modelling of continental surfaces, as well as the study of climate dynamics, are part of its field of expertise. BRGM (Bureau de recherches géologiques et minières), the French geological survey, is the public organisation of reference for the management of soil risks from a sustainable development perspective.
REFERENCES
1. CCR. Natural catastrophes in France, Assessment 1982- 2020. (2020).
2. Gourdier, S. & Plat, E. Impact du changement climatique sur la sinistralité due au retrait-gonflement des argiles. in Journées Nationales de Géotechnique et Géologie de l’Ingénieur (JNGG) (2018).
3. Noilhan, J. & Mahfouf, J. F. The ISBA land surface parameterisation scheme. Glob. Planet. Change 13, 145–159 (1996).
4. Boone, A., Masson, V., Meyers, T. & Noilhan, J. The influence of the inclusion of soil freezing on simulations by a soil-vegetation-atmosphere transfer scheme. J. Appl. Meteorol. 39, 1544–1569 (2000).
5. Kendall, M. G. Rank Correlation Methods. (Griffin London, 1970).
6. ECMWF. Soil wetness index calculation - Copernicus Services - ECMWF Confluence Wiki. https://confluence. ecmwf.int/display/COPSRV/ Soil+wetness+index+calculation (2021).
7. Kapsambelis, D., Moncoulon, D. & Cordier, J. An innovative damage model for crop insurance, combining two hazards into a single climatic index. Climate 7, (2019).
CITATION
Barthelemy et al, Understanding the phenomenon of clay shrinking and swelling through an indicator aggregated to the commune: the magnitude of droughts. In CCR 2022 Scientific Report; CCR, Paris, France, 2022, pp. 20-22