Protected areas are great tools to conserve the world's declining biodiversity. Often located in populated areas, protected areas are however increasingly expected to provide other services to that of biodiversity conservation, such as delivering ecosystem services and supporting local livelihoods. Unfortunately, the lack of data available on the distribution of biodiversity and ecosystem services, particularly in biodiverse but remote tropical protected areas, hinders the design and implementation of effective management strategies necessary to ensure these areas fulfill their potential.

Passionate about conservation and eager to help biodiversity, I embarked six years ago on a PhD project at the University of Antwerp aiming to design science-based, effective management recommendations for the Tunari National Park, a data-poor protected area located in the Andes of Bolivia. The Tunari National Park, and especially its Southern Slope, is an interesting case-study area to carry out this type of work: it is one of the strongholds for high-altitude Polylepis forests, an important habitat for rare and threatened species, which also provides numerous key ecosystem services to the city of Cochabamba located downstream. However, due to the lack of an integrated management strategy in the area, forest fragments are quickly converted to pastures, agricultural fields and exotic tree stands and it is feared they will soon disappear.

In our study, we combined local knowledge, data collected in the field, species distribution modelling tools with the systematic conservation planning tool Marxan with Zones to create six land use plans ensuring different levels of biodiversity conservation while minimizing the opportunity cost associated with the loss of exploitable land to local communities. Using four water-related ecosystem services modelled with the AguAAndes tool (freely available at www.policysupport.org), we then identified the potential trade-offs between ecosystem services delivery and the different land use plans we created.

We find that conserving most of the avifauna associated with Polylepis forests would require the strict protection of a little less than 15% of the Southern Slope, leaving the rest of the area available for agricultural, pastoral and forestry activities. There is a synergy between biodiversity conservation and three of the services we modelled: levels of water stress, water pollution and water runoff decrease across the Southern Slope as more Polylepis forest fragments and suitable bird habitats are conserved in the land use plans. As conservation areas expand however, the displacement of agricultural activities to erosion-prone, steeper areas results in an increase of the total amount of soil erosion across the Southern Slope. While protecting all remaining Polylepis fragments is crucial to ensure ecosystem services delivery on the Southern Slope, we suggest exploring the potential for agroforestry activities to improve the delivery of ecosystem services, especially on steeper slopes.

As the last chapter of my thesis, this paper is the culmination of years of work to bring data together and learn many new techniques. Seeing this paper published is therefore very rewarding and I am glad to be given the opportunity to share my work with the scientific community. I hope this study will encourage scientists and managers to identify potential trade-offs between biodiversity conservation and ecosystem services delivery using a similar framework, and will ultimately contribute to improve conservation in protected areas.