SUPPORTING INFORMATION
Predicting the future effectiveness of protected areas for bird conservation in Mediterranean ecosystems under climate change and novel fire regime scenarios
APPENDIX S2: Detailed information on land cover scenarios: storylines and simulations.
In this appendix we provide a detailed description of landscape simulations and the storylines underlying the potential future land cover pathways in Mediterranean-type ecosystems. These storylines are based on the key socio-ecological driving forces that have the potential to affect landscape dynamics in the study region, such as climate change, fire disturbance regime, large-scale fire management, land abandonment and forest harvesting for bioenergy uses (see also Regos et al. 2015).
Storylines
The description of four regional storylines developed for Catalonia is supported by previous collaborative works wherein conservation ecologists, forest engineers and technical fire brigades (GRAF) were involved (Brotons et al. 2013; Regos et al. 2014):
1) Stop all fires
Increased fire risk, higher burning frequencies and larger burnt areas are expected in the Western Mediterranean region due to vegetation encroachment following land abandonment, coupled with increasingly severe weather. Current fire management option based on suppressing “all fires” is characterized by high fire suppression levels regardless of the climatic severity of the year. It is expected that this strategy will continue to be applied in a near future.
2) No suppression
Nowadays, firefighters have become more efficient at suppressing small and low-intensity fires, but the number of large and high-intensity fires has considerably increased during the last decades. According to “fire paradox” (Minninch, 1983) large fires are a modern artefact induced by successful fire exclusion along the 20th century. Assuming a paradigm shift in fire management, we envisaged another set of baseline treatments embodying the ‘no fire suppression’ strategy as a counterpoint to the current trend.
3) Let-burn
Rural communities have traditionally used fire in rangeland management and agriculture, as well as in forests, and the long fire history in the Mediterranean region has created ecosystems that need fire for their sustainability. An opportunistic fire suppression strategy based on reducing active firefighting efforts in controlled “mild” weather conditions provides further firefighting opportunities in adverse years (Houtman et al., 2013; Regos et al., 2014). Here it is considered as a possible pathway that the fire management policies might follow in the next decades.
4) Forest biomass extraction
The biomass harvesting as a fuel could reduce energy consumption in local communities, among other socioeconomic and environmental implications at regional level (Mason et al., 2006; Becker et al., 2009; Evans & Finkral, 2009; Abbas et al., 2011). In Catalonia, a forest harvesting strategy has been recently approved and includes specific targets for biomass-derived energy (GENCAT, 2014). Its effectiveness of this fuel-reduction treatment for suppressing wildfires has been recently shown in Catalonia (Regos et al. submitted). We designed a set of scenarios characterized by forest harvesting in optimal areas from a logistic and economic viewpoint (i.e. favourable site conditions with gently slopes and small extraction distances) assuming thus a cost-effective forestry biomass harvesting. An additional set of scenarios was designed to represent an environmental perspective wherein biomass extraction is prohibited in protected areas.
Scenario design and implementation can be found in Regos et al. (2015).
Landscape simulations
We used the MEDFIRE model to simulate future land cover changes derived from spatial interactions among fire regime, vegetation dynamics and fire management policies (Brotons et al., 2013; De Cáceres et al., 2013; Regos et al., 2014). MEDFIRE is a spatially explicit dynamic fire-succession model designed to integrate climate and anthropogenic drivers. It allows examining their combined effect on fire regime and, in turn, on land cover at short- and medium-term time scales in a Mediterranean context. MEDFIRE is based on observed time series to simulate the future effect of primary processes driving vegetation dynamics (i.e. natural succession, post-fire regeneration and maturation processes) and fire regime (i.e. fire ignition, fire spread, fire suppression and fire effects) in the landscape. In the MEDFIRE model, fires are simulated until the potential annual area to be burnt is reached. Potential annual area refers to the area that is expected to burn according to the historical fire data (1975–99 period). According to previous research (Piñol et al., 1998), climatically adverse years are characterized by a high number of weather risk days (‘adverse years’), as opposed to years dominated by mild weather conditions (‘mild years’). Thus, potential burnt area and fire size distributions depend on the climatic severity of the year: (1) the probability of a year being adverse increases from 0.30 to 0.59 for time-slice 2050 in A2 emission scenarios; and from 0.30 to 0.62 for time-slice 2050 in B2 emission scenarios (more details in Regos et al., 2015). Vegetation encroachment due to land abandonment (hereafter land abandonment) is explicitly integrated into MEDFIRE to simulate the succession from abandoned open land (i.e. shrubland) to forest and its interaction with fire regime. Land cover in 2000 was represented by means of two raster layers at 100-m resolution: land cover type (LCT) and time since last fire (TSF). In particular, the model assumes that forest cover types are relatively stable, so a type-conversion can only occur after burning. Succession without burning can occur only from shrubland to forest. This land cover change takes place depending on the availability of mature forest in neighbouring cells and the TSF of shrubland that will potentially change. Post-fire transitions in dominant species are implemented according to two approaches: non-spatial stochastic transitions or neighbourhood species contagion.
To deal with the stochastic nature of wildfires, the land cover layers were then simulated 10 times (hereafter runs) for 2050 using MEDFIRE model under the different combinations of 6 fire management scenarios and 2 climate change scenarios (Table 1, Appendix S1 and Regos et al. 2015 for the detailed description of the scenarios). To describe predicted vegetation changes under each scenario, we used the outputs of the simulation runs and we calculated the area occupied by each land cover type. Some land cover types do not influence fire dynamics (i.e. water, rocks and urban areas), whereas farmland was assumed to be static but to allow fire to spread through it. Fire can affect farmlands but they do not directly shift to other habitat types after fire unless an additional land use change occurs. A transition from farmland to shrubland requires a land use change that was not simulated in the context of the present study. Detailed estimates of future land cover were obtained considering the following categories: (1) coniferous tree species (mainly dominated by Pinus halepensis, Pinus nigra, Pinus pinea, Pinus sylvestris) (2) deciduous tree species (Quercus ilex and Quercus suber), (3) shrubland and (4) farmland (represented by different types of cropland). Additional variables were measured to represent other fire-mediated land cover properties such as the vertical structure or the maturation of the vegetation by calculating the coverage of three different age-classes of the vegetation: (5) older vegetation (>30 years since fire), (6) mid-age vegetation (10–30 years since fire), and (7) recently burned vegetation (<10 years since fire). To match bird data resolution in Catalonia, we calculated the area covered by each variable within 1-km resolution squares.
References
Abbas D, Current D, Ryans M, Taff S, Hoganson H, Brooks KN (2011) Harvesting forest biomass for energy – An alternative to conventional fuel treatments: Trials in the Superior National Forest, USA. Biomass and Bioenergy, 35, 4557–4564.
Becker DR, Larson D, Lowell EC (2009) Financial considerations of policy options to enhance biomass utilization for reducing wildfire hazards. Forest Policy and Economics, 11, 628–635.
Brotons L, Aquilué N, de Cáceres M, Fortin M-J, Fall A (2013) How fire history, fire suppression practices and climate change affect wildfire regimes in Mediterranean landscapes (ed Bohrer G). PLoS ONE, 8, e62392.
De Cáceres M, Brotons L, Aquilué N, Fortin M-J (2013) The combined effects of land-use legacies and novel fire regimes on bird distributions in the Mediterranean (ed Pearman P). Journal of Biogeography, 40, 1535–1547.
Evans AM, Finkral AJ (2009) From renewable energy to fire risk reduction: a synthesis of biomass harvesting and utilization case studies in US forests. GCB Bioenergy, 1, 211–219.
GENCAT (2014) Estratègia per promoure l’aprofitament energètic de la biomassa forestal i agrícola. 106 pp.
Houtman RM, Montgomery CA, Gagnon AR, Calkin DE, Dietterich TG, Mcgregor S (2013) Allowing a wildfire to burn: estimating the effect on future fire suppression costs. International Journal of Wildland Fire, 22, 871–882.
Mason CL, Lippke BR, Zobrist KW et al. (2006) Investments in fuel removals to avoid forest fires result in substantial benefits. Journal of Forestry, 104, 27–31.
Minninch RA (1983) Fire Mosaics in Southern California and Northern Baja California. Science, 219, 1287–1294.
Piñol J, Terradas J, Lloret F (1998) Climate warming, wildfire hazard, and wildfire occurrence in coastal eastern Spain. Climatic Change, 38, 345–357.
Regos A, Aquilué N, Retana J, De Cáceres M, Brotons L (2014) Using unplanned fires to help suppressing future large fires in Mediterranean forests (ed Añel JA). PLoS ONE, 9, e94906.
Regos A, D’Amen M, Herrando S, Guisan A, Brotons L (2015) Fire management, climate change and their interacting effects on birds in complex Mediterranean landscapes: dynamic distribution modelling of an early-successional species — the near-threatened Dartford Warbler (Sylvia undata). Journal of Ornithology.