Dynamic Data Driven Ensemble for Wildfire Behaviour Assessment: A Case Study

Abstract

Abstract. Wildfire information has long been collected in Europe, with particular focus on forest fires. The European Forest Fire Information System (EFFIS) of the European Commission complements and harmonises the information collected by member countries and covers the forest fire management cycle. This latter ranges from forest fire preparedness to post-fire impact analysis. However, predicting and simulating fire event dynamics requires the integrated modelling of several sources of uncertainty. Here we present a case study of a novel conceptualization based on a Semantic Array Programming (SemAP) application of the Dynamic Data Driven Application Systems (DDDAS) concept. The case study is based on a new architecture for adaptive and robust modelling of wildfire behaviour. It focuses on the module for simulating wildfire dynamics under fire control scenarios. Rapid assessment of the involved impact due to carbon emission and potential soil erosion is also shown. Uncertainty is assessed by ensembling an array of simulations which consider the uncertainty in meteorology, fuel, software modules. The event under investigation is a major wildfire occurred in 2012, widely reported as one of the worst in the Valencia region, Spain. The inherent data, modelling and software uncertainty are discussed and preliminary results of the robust data-driven ensemble application are presented. The case study suitably illustrates a typical modelling context in many European areas -for which timely collecting accurate local information on vegetation, fuel, humidity, wind fields is not feasible -where robust and flexible approaches may prove as a viable modelling strategy.

Modelling soil erosion at European scale: towards harmonization and reproducibility

“…Furthermore, approximated rapid assessments of the impact of disturbances (e.g. wildfires, de Rigo et al, 2013a;Di Leo et al, 2013) may be performed by exploiting the RUSLE modular architecture which easily allows potential and actual erosion rates to be estimated for different environmental conditions by simply considering different arrays of layers. The further seamless integration of multiple estimates in the erosivity component (see Sect.…”

“…It is also true that concentrated surface runoff and gully erosion may occur within depressions which could also be old landslide scars (Valentin et al, 2005;Mazaeva et al, 2013). Forest fires can also significantly affect soil erosion (Terranova et al, 2009;Di Leo et al, 2013). Their effects on the catchment hydrology may cause an increasing of soil erosion by water (Vafeidis et al, 2007).…”

“…Their effects on the catchment hydrology may cause an increasing of soil erosion by water (Vafeidis et al, 2007). Wildfires can totally remove the protective surface layers (litter and vegetation cover) and increase the hydrophobic condition at or below the ground surface with a consequent increase of the soil erosion rate (Di Leo et al, 2013). Under these circumstances, the mentioned natural hazards might be interlinked with soil erosion rates even close to the maximum potential erosion (i.e.…”

Modelling soil erosion at European scale: towards harmonization and reproducibility

“…It is also true that concentrated surface runoff and gully erosion may occur within depressions which could also be old landslide scars (Valentin et al, 2005;Mazaeva et al, 2013). Forest fires can also significantly affect soil erosion (Terranova et al, 2009;Di Leo et al, 2013). Their effects on the catchment hydrology may cause an increasing of soil erosion by water (Vafeidis et al, 2007).…”

Estimating the effects of water-induced shallow landslides on soil erosion

“…The proposed modelling approach is also suitable to be applied in applications having a wider spatial extent and to be potentially implemented in a transdisciplinary context. For example, the relevant effect of wildfires on soil erosion and landslide susceptibility [46,47] could be modelled with a higher reliability integrating the proposed approach. As stated in de Rigo et al [47], wildfires can considerably increase soil erosion by water and landslide susceptibility.…”