Effect of prescribed burning on the mitigation of fire risk was discussed in fire prone ecosystems of three European countries (Greece, France and Portugal). Two study cases were presented using two different fire modelling approaches, the spatial fire growth model FARSITE and FIRETEC a physically based fire propagation model. This study showed that fuel management with prescribed burning had a significant effect on fire propagation hazard. It needs to be completed with further studies on prescribed burning effects on the other fire risk components such as ignition likelihood or vulnerability (fire effects on trees). First results also showed the need of optimizing the spatial pattern of fire application both when applied extensively on the landscape, and when located on a fuel break network. The first study case concerned a conifer forest located in Kassandra peninsula in northern Greece, which has been affected by a 7120 ha wildfire in August 2006. Prescribed burning effectiveness was assessed by using the FARSITE simulator for comparing fire spread over the landscape before and after prescribed burning application on a network of stripe-shaped fuel-breaks. Analysis showed a significant reduction of fire potential (fireline intensity, flame length and total burned area) as a result of fuel treatments. The second study case concerned a shrubland and woodland mosaic in limestone Provence, France, where prescribed burning has been used as a fuel reduction tool for more than 15 years The effectiveness of prescribed burning was assessed by comparing fire potential (rate of spread, fireline intensity) before and in the years following prescribed burning treatments. Analysis carried out at the stand scale gave valuable information on the immediate potential fire hazard reduction after prescribed burning. In order to estimate the appropriate prescribed burning return interval, two management scenarios were used (i) a fire intensity threshold compatible with efficient and safe fire fighting or (ii) a shrub encroachment threshold easy to assess by forest managers. In both cases, the prescribed burning return interval varied from 2 to 3 years, depending on fuel types. When applying prescribed burning on a 15 years period, fire risk mapping on the whole State forest showed a higher fire risk reduction when using the fire intensity threshold than a fuel encroachment threshold. A third study case in the Portuguese mainland tested an important assumption of the fire paradox: does fuel age have an influence on burn probability and wildfire size? Fire frequency analysis of mapped fire perimeters ≥10 ha (1998-2008) indicated that fire occurrence was weakly to moderately dependent of fuel age. However, contrarily to the usual expectation, age-dependency did not decrease for larger fire size classes (i.e with increasingly extreme fire weather). Finally, we explored the variation in fire size statistics with fuel age. Older fuels (roughly >8 years after burning) were associated to more variable fire sizes, allowing much larger wildfires. From the results we infer that landscapes submitted to a mean prescribed fire return interval of 4 years will not support wildfires larger than 500 ha. However, adequate spatial planning of the treatments is expected to reduce the annual treatment rate. This deliverable contributes to solving the Fire Paradox because it clearly shows how fuel modification can decrease wildfire hazard and severity which is the first specific objective to be addressed in Integrated Wildland Fire Management. The state of the art chapter has shown that several fuel modification techniques, such as mechanical treatments, are available and used in Europe, but it has also demonstrated that prescribed burning although being increasingly used in some countries, still has to be strongly supported with the objective of the wisest complementarities between fuel treatments. Also, prescribed burning is the approach based on the paradox of fire, reducing fuels strategically before they can be used by wildfires, thus decreasing wildfire risk and severity