The Fate of Priority Areas for Conservation in Protected Areas: A Fine-Scale Markov Chain Approach

Abstract

Park managers in alpine areas must deal with the increase in forest coverage that has been observed in most European mountain areas, where traditional farming and agricultural practices have been abandoned. The aim of this study is to develop a fine-scale model of a broad area to support the managers of Paneveggio Nature Park (Italy) in conservation planning by focusing on the fate of priority areas for conservation in the next 50-100 years. GIS analyses were performed to assess the afforestation dynamic over time using two historical maps (from 1859 and 1936) and a series of aerial photographs and ortho-photos (taken from 1954 to 2006) covering a time span of 150 years. The results show an increase in the forest surface area of about 35%. Additionally, the forest became progressively more compact and less fragmented, with a consequent loss of ecotones and open habitats that are important for biodiversity. Markov chain-cellular automata models were used to project future changes, evaluating the effects on a habitat scale. Simulations show that some habitats defined as priority by the EU Habitat Directive will be compromised by the forest expansion by 2050 and suffer a consistent loss by 2100. This protocol, applied to other areas, can be used for designing long-term management measures with a focus on habitats where conservation status is at risk.

Simulating Forest Cover Changes of Bannerghatta National Park Based on a CA-Markov Model: A Remote Sensing Approach

“…64 countries have less than 10% forest cover and of these ten have no forest at all. Globally, protected areas (PAs) still remain the most popular conservation approach to protect biodiversity [6][7][8][9][10][11][12][13][14][15]. Often it is assumed that legal protection will bring positive impact on the forest cover of an area and help in lowering deforestation [16].…”

“…In other words, to understand the impact of a protection, we want to know what would have happened to forest cover if there would have been no policy intervention, i.e., would the rate of change in forest cover have differed under differing political or institutional frameworks. Additionally, policy interventions within PAs have varied between and within countries, which make any generalization about the effectiveness of PAs even harder and necessitate case-by-case evaluation [12][13][14][15].…”

Use of cellular automata in the study of variables involved in land use changes

“…This takes into account the error in the classification of the land use maps." (Tattoni et al 2011). It is often set at 0.15-0.10 because a common value of accuracy for a land use map is 85 %-90 %.…”

Toward Innovative Integrated Approaches for the Conservation of Mammals

“…Further, using available information of historical habitat fragmentation and current impacts to ecosystems (e.g. trends in agriculture and cattle ranching expansion, road implementation), Markov Chain models could be applied to derive future spatially explicit models of habitat loss (see Tattoni et al 2011). These models could be than coupled to species distribution models to 'correct' possible biases in potential distributions and be evaluated under current IUCN thresholds to assign species extinction risk.…”