This study investigates the role of prescribed grazing as a sustainable fire prevention strategy in Mediterranean ecosystems, with a focus on Sardinia, an area highly susceptible to wildfires. Using FlamMap simulation software, we modeled fire behavior across various grazing and environmental conditions to assess the impact of grazing on fire severity indicators such as flame length, rate of spread, and fireline intensity. Results demonstrate that grazing can reduce fire severity by decreasing combustible biomass, achieving reductions of 25.9% in fire extent in wet years, 60.9% in median years, and 45.8% in dry years. Grazed areas exhibited significantly lower fire intensity, particularly under high canopy cover. These findings support the integration of grazing into fire management policies, highlighting its efficacy as a nature-based solution. However, the study’s scope is limited to small biomass fuels (1-h fuels); future research should extend to larger fuel classes to enhance the generalizability of prescribed grazing as a fire mitigation tool.

Estimating timber volume in forests is crucial for effective management and commercial purposes, particularly for forest owners and managers. Accurate volume estimates inform management decisions, growth monitoring, and silvicultural treatments. Traditional methods often involve destructive sampling, which presents logistical and economic challenges. This study evaluates the non-destructive methods of Hossfeld’s, centroid sampling, and Pressler’s to estimate the volume of 42 Pinus pinaster and 50 Pinus radiata standing trees. For each species, the estimated volume was then compared with that calculated using precise measurements on felled trees and with the volume calculated using the simplified method F = 0.5. For both species, Pressler’s method showed the highest accuracy, with RMSE values of 28 and 102 dm³ for P. pinaster and P. radiata, respectively. However, the centroid and Hossfeld’s methods also showed satisfactory results: for P. pinaster the RMSE was 40 and 47 dm³, while for P. radiata it was 100 and 118 dm3, respectively. These estimates resulted largely more accurate than those obtained using the F=0.5 method (60 dm³ for P. pinaster and 432 dm³ for P. radiata). All the quick methods tested with F=0.5 show commendable accuracy, with Pressler’s method yielding the best results. According to the Percent Relative Bias, all the three methods slightly tend to underestimate the volume. Our results support the viability of non-destructive methods for rapid volume estimation without extensive felling and measurement. However, operational application in the field and validation across different tree species and growth habits are recommended. These methods offer viable alternatives to traditional approaches, reducing resource intensity and facilitating more frequent and detailed forest surveys. Our findings support the application of non-destructive techniques in forest management, potentially streamlining volume estimation and improving the accuracy of forest inventories.