the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 31 Mar 2026
A Joint Space-Time Probabilistic Model for Agricultural Droughts, Hydrological Droughts and Fire Weather in France
Abstract. Agricultural droughts, hydrological droughts and wildfires have significant environmental and socioeconomic consequences. These hazards are physically linked because they share a number of forcings, and their space-time properties are important as impacts result from their spatial extent and duration, in addition to their intensity. This paper introduces a probabilistic model adapted to the description of multiple spatial hazards, based on the combination of simple ingredients: regressions to describe dependencies between hazards, principal component analysis to describe spatial dependence, and simple covariance functions to describe time dependence and residual spatial dependence. This results in a modular framework that decomposes a complex model into several simpler models. This model is then used to analyze the observed Soil Wetness Index, Fire Weather Index and river flows in France over the last six decades, and in particular to estimate the probability of occurrence of the remarkable 2022 summer event. Locally, the 2022 summer was extreme in terms of agricultural drought over a large part of the country, but was rather moderate in terms of hydrological drought and fire weather. However, the magnitude, spatial extent and duration of the event become extreme nearly everywhere when the three hazards are considered together. The underlying trends affecting all three hazards have more than doubled the probability of the event during the historical period, and future projections suggest that it might become common by the end of the century with global warming.
- Preprint
(1909 KB) - Metadata XML
-
Supplement
(10649 KB) - BibTeX
- EndNote
Status: open (until 30 Jun 2026)
- RC1: 'Comment on egusphere-2026-1406', Anonymous Referee #1, 22 May 2026 reply
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 316 | 95 | 21 | 432 | 430 | 22 | 26 |
- HTML: 316
- PDF: 95
- XML: 21
- Total: 432
- Supplement: 430
- BibTeX: 22
- EndNote: 26
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
This paper analyzes the impact of low soil moisture index, fire state, and droughts in France using six decades of observations and compares this with the record-breaking summer event of 2022 in terms of magnitude and spatial extent. Using a suite of GCM-forced RCM combinations, the study finds an increased likelihood of events similar in magnitude and extent to the 2022 event under future conditions, while accounting for the interdependence among all three drivers.
Although the manuscript addresses a scientifically relevant topic in the context of a warming climate, the writing is difficult to follow in several places and would benefit from improved clarity and organization. Furthermore, the methodology adopted for drought identification and its relationship with hydroclimatic variables appears to be incorrectly defined and requires substantial revision. The specific comments are provided below. Therefore, I recommend a major revision.
L (Line number in this review)
1) L60: Annual Exceedance Probability?
2) In Data: At which depth is the soil wetness value derived? This information is important because deep soil moisture and surface soil moisture play fundamentally different roles in the hydrological cycle. Deep soil moisture is critical for sustaining streamflow baseflow, whereas surface soil moisture directly influences agricultural production by supplying water to the root zone. The subsurface storage reduction and groundwater-related drought responses are generally slower and weaker than streamflow related responses (Jin et al., 2026).
The reader can also benefit if the study area map with elevation and locations of stream gauges with respect to climatology (mean temperature and annual average rainfall pattern) of the study area are shown. Right now, it is not clear, where the density of stations is sparse/high and how the regional climatology may influence the likelihood of compound extremes. This would also help to assess the tole of causal interactions in mediating spatial dependence, driving compound hazards.
3) In Methods:
On a similar token, in L127, while the correlation between FWI and Q is shown as a consequence of SWI as a common driver, here lies the role of soil moisture sampling, whether it is derived from surface level, which influences fire hazard or sub-surface level, impacting the baseflow responses, in turn influencing water availability in streams or lakes. For example, in L184, if we see physically, there will be little association between FWI and SWI anomalies, while a lagged effect can be possible. Therefore, a lagged interactions can be evaluated to show if FWI anomalies are conditioning SWI, the response variable.
4) L133: Sentence is too complex to easily assimilate by reader.
5) L158: Please clarify whether the polynomial function in Eq. 3 is calculated centered wrt time or in space.
6) L164: The rationale for applying a Gaussian copula after PCA is unclear. PCA already reduces dimensionality and removes much of the interdependence structure among variables through orthogonal transformation. Therefore, the additional benefit of imposing a Gaussian copula framework is not sufficiently justified in the manuscript.
7) More importantly, Gaussian copulas are known to inadequately represent tail dependence and joint extremes, because they assume symmetric dependence and weak tail connectivity. This limitation is particularly critical when analyzing hydroclimatic extremes and compound drought events, where tail behavior governs risk propagation and severity.
8) Page 8, L177: The model is pretty complex, although Fig. 2 shows regression terms, the dependent and independent variables, i.e., who is regressing with what is not explained properly. A simplified flowchart explaining overall framework highlighting each process flow and variable would be helpful.
9) Page 12: L267, ‘Groundwater-dominated northern regions’ appears speculative unless supported by hydrological evidence. The interpretation could be strengthened by linking the observed behavior to documented baseflow contributions or baseflow index characteristics, since sustained groundwater-fed baseflow strongly influences streamflow persistence and drought propagation.
10) Figure 8 Caption: In this figure contours with yellow shade are superimposed by violate shade. Please provide legends and show these contours represent probability of which set of variables.
11) Figure 9 Caption: ‘Lines correspond to the cases of independence’ is somewhat ambiguous. Since the probabilities appear to be derived by multiplying the marginal probabilities of each variable, it would be more precise to state: ‘Lines correspond to the joint probability under the assumption of independence.
12) L330: This could be deemed and highlighted as a potential caveat of the current analyses as interdependencies between drivers can alter in future time windows (Zscheischler et al., 2018).
References
Jin, L., Xue, H., Ma, G., A, Y., Wang, G., Li, K., Wu, J., Wang, Y., Xue, B., 2026. Propagation of meteorological drought to streamflow and baseflow droughts using an improved SWAT model. Journal of Hydrology: Regional Studies 66, 103538. https://doi.org/10.1016/j.ejrh.2026.103538
Zscheischler, J., Westra, S., Hurk, B.J., Seneviratne, S.I., Ward, P.J., Pitman, A., AghaKouchak, A., Bresch, D.N., Leonard, M., Wahl, T., 2018. Future climate risk from compound events. Nature Climate Change 8, 469-477.