The use of digital aerial photography as support for restoration, management and habitat monitoring programmes

The use of digital aerial photography as support for restoration, management and habitat monitoring programmes João Filipe Airosa Gomes Mestrado em Ecologia, Ambiente e Território Departamento de Biologia 2012 Orientador Professor Doutor João Honrado, Faculdade de Ciências da Universidade do Porto Coorientador Professor Doutor Renato Henriques, Universidade do Minho FCUP The use of digital aerial photography as support for restoration, management and habitat monitoring programmes 2 Acknowledgements I would like to express my deep gratitude to Professor João Honrado and Professor Renato Henriques, my research supervisors, for their guidance, encouragement and useful critiques of this research work. I would also like to thank Dr. Gabriela Santos for the support on the Turfeira da Almofrela restoration project, Dr. Sofia Vaz for the analysis of the data obtained from the Praia da Aguçadoura study and Dr. Paulo Alves for his help with the Serra de Arga habitat mapping. I would also like to extend my thanks to the technicians of the ICETA laboratory for their kindness in welcoming me into their workspace. At last, but definitely not least, I wish to thank my family for their support and encouragement throughout my study. FCUP The use of digital aerial photography as support for restoration, management and habitat monitoring programmes 3 Summary With the rapid loss of biodiversity we face nowadays, urgent action needs to be taken in order to stop it, or at least to reduce it. Habitat restoration, new management approaches and monitoring programmes are some of the best ways to do so. However, these initiatives are often not taken seriously by scientists and stakeholders, and sometimes they are seen as a waste of time or too much work for very little gain. Ecological assessment and monitoring is intrinsically spatially-explicit, and imagery from airborne devices is an important source of information. In order to make data acquisition easier and faster, we used a digital camera coupled to a supporting structure built to hang on an inflatable kite or a pole (close-range photogrammetry). The resulting photos were transferred into a computer, which is then used generate a three-dimensional digital terrain model (3D DTM) and an orthophotomap. Then, by means of visual interpretation on a Geographical Information System (GIS) software, depending on the study area, different approaches were taken to extract ecologically relevant information. This way, several maps were obtained, each of them being an example of how this method can have a wide array of applications: detailed habitat mapping to support active management, an ecological restoration plan proposal for a fine-grained habitat mosaic, and a sampling grid where the spatial patterns of individuals of different plant species were identified and studied statistically. These high-quality cartographic results were obtained in a rather easy and fast way, meaning that this method can save time, effort and money, and might therefore be the answer to some of the problems with ecological research nowadays. Key-words Biodiversity loss, DTM, GIS, Natura 2000, orthophotomap, photointerpretation, priority habitat, remote sensing FCUP The use of digital aerial photography as support for restoration, management and habitat monitoring programmes 4 Resumo Devido à acelerada perda de biodiversidade que enfrentamos hoje em dia, é necessário tomar medidas para parar ou pelo menos reduzir essa perda. Restauro ecológico, novas medidas de gestão e programas de monitorização são das melhores ferramentas que temos para o conseguir. Infelizmente, estas iniciativas não são levadas a sério pelos cientistas e outros intervenientes, visto que muitas vezes são vistas como uma perda de tempo, requerem esforço a mais e dão resultados pouco úteis. Monitorização e avaliações ecológicas precisam necessariamente de informação espacial, e, como tal, imagens aéreas são importantes fontes de informação. Com vista a tornar a aquisição de dados mais simples e mais rápida, utilizou-se uma máquina fotográfica digital montada num suporte construído para ser suspenso num papagaio ou numa cana (fotogrametria digital à curta distância). As fotografias obtidas foram transferidas para um computador que gera modelos digitais de terreno tridimensionais e ortofotomapas. Posteriormente, dependendo da área de estudo em questão e com recurso à interpretação visual das imagens obtidas num ambiente de Sistema de Informação Geográfica (SIG), fizeram-se várias análises com vista a extrair informações ecologicamente relevantes. Deste modo, obtiveram-se vários mapas, cada um constituindo um exemplo das várias aplicações que este método pode ter: cartografia detalhada de habitats para apoiar a gestão activa da área, um plano de restauro ecológico para um local onde se encontra um mosaico de habitats muito fino e uma grelha de amostragem para a análise dos padrões espaciais de indivíduos de espécies vegetais diferentes, onde foram efetuadas análises estatísticas. Estas cartografias de alta-resolução foram todas obtidas de um modo mais simples e rápido do que o seriam caso fossem utilizadas técnicas mais tradicionais que se usam actualmente, o que significa que este método pode poupar tempo, esforço e dinheiro, constituindo assim uma solução para alguns dos problemas dos estudos ecológicos nos tempos que correm. Palavras-chave Deteção remota, fotointerpretação, habitat prioritário, MDT, Natura 2000, ortofotomapa, perda de biodiversidade, SIG FCUP The use of digital aerial photography as support for restoration, management and habitat monitoring programmes 5 Index Acknowledgements ......................................................................................................................... 2 Summary ........................................................................................................................................ 3 Resumo .......................................................................................................................................... 4 List of tables and figures ................................................................................................................. 7 List of abbreviations ........................................................................................................................ 8 1. Introduction ................................................................................................................................. 9 1.1. Biodiversity loss .................................................................................................................... 9 1.2. Scope and key issues of ecological monitoring ................................................................... 10 1.3. Remote sensing in ecological monitoring ............................................................................ 11 1.4. Close-range photogrammetry ............................................................................................. 12 1.5. Research objectives ........................................................................................................... 13 2. Materials and methods .............................................................................................................. 14 2.1. Acquisition and processing of close range aerial photos ..................................................... 14 2.1.1. Data acquisition ............................................................................................................... 14  Preparing the study area ............................................................................................ 14  Kite system ................................................................................................................. 14  Pole system ................................................................................................................ 15 2.1.2. Data processing ............................................................................................................... 15  DTM and orthophotomap generation .......................................................................... 15 3. Results and discussion ............................................................................................................. 17 3.1. Fine-scale monitoring of vegetation dynamics in coastal dunes .......................................... 17  Context and objectives ............................................................................................... 17  Study area and field survey ........................................................................................ 17  Raw DTM's and orthophotomaps ............................................................................... 18  Statistical analysis and results .................................................................................... 20  Discussion of results .................................................................................................. 24 3.2. Management of priority habitat types in mountain areas ..................................................... 25  Context and objectives ............................................................................................... 25  Study area .................................................................................................................. 25  Raw DTM's and orthophotomaps ............................................................................... 26 FCUP The use of digital aerial photography as support for restoration, management and habitat monitoring programmes 6  Field sampling ............................................................................................................ 29  Habitat mapping ......................................................................................................... 30  Discussion of results .................................................................................................. 35 3.3. Restoration of endangered habitat mosaics in mountain areas ........................................... 36  Context and objectives ............................................................................................... 36  Study area and field surveys ...................................................................................... 36  Raw DTM's and orthophotomaps ............................................................................... 37  Restoration plan proposal ........................................................................................... 40  Discussion of results .................................................................................................. 41 4. Conclusions .............................................................................................................................. 42 4.1. Synthesis of main results .................................................................................................... 42 4.2. Advantages and limitations ................................................................................................. 43 4.3. Future perspectives ............................................................................................................ 45 5. References ............................................................................................................................... 46 Appendix 1 - List of mentioned habitats ........................................................................................ 49 Index FCUP The use of digital aerial photography as support for restoration, management and habitat monitoring programmes 7 List of tables and figures Figure 1 – Raw orthophotomap for the general area of the Aguçadoura beach. ............................ 18 Figure 2 – Zoomed region of the raw orthophotomap of the Aguçadoura beach study area. ......... 19 Figure 3 – DTM of Aguçadoura beach area. .................................................................................. 20 Figure 4 – Sampling grid and point data for the Aguçadoura beach area. ..................................... 21 Figure 5 – Euphorbia paralias univariate O-ring statistic. .............................................................. 22 Figure 6 – Pancratium maritimum univariate O-ring statistic. ........................................................ 23 Figure 7 – Bivariate O-ring statistic with Euphorbia paralias as focal species. ............................... 23 Figure 8 – Bivariate O-ring statistic with Pancratium maritimum as focal species. ......................... 23 Figure 9 – Raw orthophotomap of Serra de Arga study area. ........................................................ 26 Figure 10 – Zoomed region of the raw orthophotomap of the Serra de Arga study area. ............... 27 Figure 11 – DTM of Serra de Arga study area. .............................................................................. 28 Figure 12 – Layout of the sampling points. .................................................................................... 29 Figure 13 – Area chosen for habitat mapping. ............................................................................... 31 Figure 14 – Zoomed layout of the area chosen for habitat mapping .............................................. 32 Figure 15 – Grid generated for reclassification. ............................................................................. 33 Figure 16 – Reclassification of the habitat mapping. ..................................................................... 34 Figure 17 – Orthophotomap of the Turfeira da Almofrela study area. ............................................ 37 Figure 18 – Zoomed region of the raw orthophotomap of the Turfeira da Almofrela study area. .... 38 Figure 19 – DTM of the Turfeira da Almofrela study area. ............................................................. 39 Figure 20 – Proposed restoration plan. ......................................................................................... 40 FCUP The use of digital aerial photography as support for restoration, management and habitat monitoring programmes 8 List of abbreviations GIS – Geographical Information System DTM – Digital Terrain Model DEM - Digital Elevation Model CRP - Close-range photogrammetry CBD - Convention on Biological Diversity GEO BON – Group for Earth Observations Biological Observation Network GCP’s – Ground Control Points DGPS – Differential Global Positioning System FCUP The use of digital aerial photography as support for restoration, management and habitat monitoring programmes 9 1. Introduction 1.1. Biodiversity loss Biodiversity loss is a serious global problem that since the Rio Summit in 1992 has been getting increased attention. This loss has been found to be a consequence of human activities and as such, during the Convention on Biological Diversity, numerous countries made a commitment to reduce biodiversity losses significantly by 2010, a goal that has not been reached. However, this is no reason to stop trying, it is necessary to halt future losses and promote the sustainable use of biodiversity (Duro et al., 2007; Walpole et al., 2009; Sullivan et al., 2011). The Habitats Directive (92/43/EEC), which together with the Birds Directive (79/409/EEC) make up the Natura 2000 network, the main framework for biodiversity conservation in the European Union, legally obliges member states to report on the conservation status of habitats listed under Annex I, where the protected habitats are listed, every six years. It becomes obvious that monitoring programmes are a necessity in order to abide by the law (Vanden Borre et al., 2011). The Group on Earth Observations Biodiversity Observation Network (GEO BON) is an example of a wide initiative that tries to promote biodiversity monitoring worldwide, and although it is a global initiative, not only European, it may help the countries in Europe to fulfil their obligation (http://www.earthobservations.org/geobon.shtml). It is not limited to habitat monitoring, as it encompasses everything from the genetic to ecosystem level and it is a way for several organizations to share their monitoring data, improving the quality, increasing the amount and making the data more readily available for policymakers, managers and all interested users (Pereira et al., 2010; Scholes et al., 2012). Along with monitoring, conservation and restoration projects are also invaluable to stop or even invert the current negative trends of biodiversity. Ecological restoration is the process through which a damaged, degraded or destroyed ecosystem is assisted in recovering (Society for Ecological Restoration International Science & Policy Working Group, 2004). However, when attention is given to increase one particular ecosystem function or service, the provision of the remaining services often suffers a decline. In cases such as agricultural ecosystems, the greatest challenge is to maintain productivity while also strengthening the provision of other ecosystem services, meaning that one of the greatest challenges of ecological restoration is to conciliate the economic, cultural and natural views of an ecosystem’s services (Rey Benayas and Bullock, 2012). FCUP The use of digital aerial photography as support for restoration, management and habitat monitoring programmes 16 geometric errors a Brown model (in Moffit & Mikhail, 1980) is used to determine all the correction parameters. To obtain a precise orthophotomap and a DTM, the obtained pictures must have enough overlap. Preferably, each picture must have more than 60% overlap with all the surrounding ones. Pictures are processed with a photogrammetric application in order to determine common points between them. The detection of common points is limited to a maximum of 40000 between each pair of pictures. Statistical algorithms are applied to optimize the collection of the best-correlated points and provide the elimination of outliers. The EXIF information contained in the pictures, such as focal length and image pixel dimensions, is used, in conjunction with the detected common points, correctly triangulated, to obtain a rough 3D point cloud. A bundle block adjustment algorithm (Triggs et al, 1999) is then used to refine the coordinates of the 3D point cloud that defines the scene geometry, as well as the relative motion and optical characteristics of the used camera. This algorithm allows the creation of a relative geometrically proportional dense 3D point cloud, obtained from the overlapping points, as well as the relative positions of the camera when each picture was captured. After this process, CGP’s obtained from the DGPS coordinates, are used to give the model internal orientation, and optimize all points to the best possible real geodesic position. CGP’s are visually identified in each picture and their xyz coordinates are declared to allow them to be tied to the correct real world position. Once the 3D point cloud is completely optimized, points are interpolated to build a continuous 3D surface, based on a mesh of triangular polygons, that defines the final geometry of the DEM. The orthophotomap is obtained by blending all the available pictures based on their common tie points, using a ―mosaic‖ algorithm and using the DEM to eliminate distortions caused by the terrain shape. In this work the following applications were used to achieve all the above tasks: Trimble Geomatics Office (for DGPS data post-processing, see http://www.trimble.com/index.aspx); and Agisoft Photoscan (for photogrammetric calculations and DEM and orthophotomap generation, see http://www.agisoft.ru/products). FCUP The use of digital aerial photography as support for restoration, management and habitat monitoring programmes 17 3. Results and discussion 3.1. Fine-scale monitoring of vegetation dynamics in coastal dunes  Context and objectives The motivations for this study result from previous research in coastal dynamics and dune vegetation (Honrado et al. 2010, Macedo et al. 2010), which has been aimed to study the vegetation responses to stress and identify robust indicators of changes in vegetation dynamics, in order to implement them in monitoring programmes to assess impacts of environmental changes and restoration programmes. Given the reduced area and in order to maximize the resolution of the model, only the pole system was used to analyse and map the test area. This is important because the goal of this project was to identify indicators based on individual plants or plant species, and some of them might be too small to distinguish even in a model generated from the kite system, thus requiring a finer resolution.  Study area and field survey The study area is a coastal dune system located in Aguçadoura beach, in Póvoa de Varzim, Oporto district, and was the smallest study area used in the present thesis. Coastal habitats are very prone to stress and disturbance since they are the interface between land and ocean, which in turn, also makes them very dynamic (Honrado et al. 2010). This particular study area only showed mild signs of human disturbance and some interesting evidences of facilitation among plant species, making it an interesting area to apply these new techniques because it showcases their potential. Field work from which this data was obtained was carried out on the 24th of July, 2012. FCUP The use of digital aerial photography as support for restoration, management and habitat monitoring programmes 18  Raw DTM's and orthophotomaps Figures 1 to 3 illustrate the main results obtained during the field survey (orthophotomaps and a digital terrain model. Many important aspects of dune ecosystems and particularly of their vegetation can be easily observed. Figure 1 – Raw orthophotomap for the general area of the Aguçadoura beach. FCUP The use of digital aerial photography as support for restoration, management and habitat monitoring programmes 19 Figure 2 – Zoomed region of the raw orthophotomap of the Aguçadoura beach study area. FCUP The use of digital aerial photography as support for restoration, management and habitat monitoring programmes 20 Figure 3 – DTM of Aguçadoura beach area.  Statistical analysis and results A 10 by 5 meter grid with 1 by 1 meter squares was generated using the ―Create fishnet‖ tool in ArcMap 10.1, and flowers of species Pancratium maritimum and Euphorbia paralias were marked in different shapefiles by means of visual interpretation (Figure 4). FCUP The use of digital aerial photography as support for restoration, management and habitat monitoring programmes 21 Figure 4 – Sampling grid and point data for the Aguçadoura beach area. Then, using the ―Point Proximity‖ tool, selecting the input features as the points in the Euphorbia shapefile, distances to each point of the Pancratium shapefile were calculated. Using these data, as well as the coordinates of each point in each shapefile, the corresponding attribute tables were exported to Microsoft Excel 2010. Spatial point pattern analysis was then conducted using O-ring statistic measures, through the mark-correlation g(r) function (g-function). The univariate g-function measures the expected number of neighbourhood points in a ring of radius r centred in an arbitrary point (i.e. the focal point, which is not counted) divided by the overall intensity of the pattern. The bivariate g-function is an extension of the first, comprising type 1 and type 2 points (where type 1 and type 2 points represent the patterns of two different species) FCUP The use of digital aerial photography as support for restoration, management and habitat monitoring programmes 22 (Wiegand and Moloney, 2004). The maximum value of the radius was set to 25 decimetres, representing an acceptable Euphorbia seed dispersion distance (Baiges et al., 1991), since the reproductive process of Pancratium is mainly vegetative. The width of the ring was set to 1 in order to avoid the jagged effect. Point-point distances were converted to decimetres in order to facilitate the visualization and interpretation of the results. Statistical significance of the results was achieved by testing the real g-functions against 199 functions simulated by null models. The 5th highest and lowest values for the generated gfunctions were used to construct simulation envelopes: if the real g(r) was smaller or greater at a given scale r than the 5th lowest or higher values, the species was regarded as having a less, or more, diverse local neighbourhood at scale r than expected by the null model (with a 95% confidence interval), respectively. As such, the univariate g-function form was interpreted as a measure of species assembly at the individual scale, i.e., intraspecific relations for each individual species: Euphorbia paralias and Pancratium maritimum. The bivariate form was applied in order to detect interspecific assembly between both species considering (a) Euphorbia as a focal species (fixed pattern) and Pancratium as a neighbourhood species (randomized pattern), and (b) Pancratium as a focal species (fixed pattern) and Euphorbia as a neighbourhood species (randomized pattern). Due to the un-homogeneous pattern of the species (Figure 4) the heterogeneous Poisson process was chosen for the null models. All analyses were performed using the software Programita (Wiegand and Moloney, 2004) and details on the estimators of the g-functions and edge correction formulas can be found in Wiegand and Moloney (2004). The results are summarized in Figures 5 to 8. Figure 5 – Euphorbia paralias univariate O-ring statistic. FCUP The use of digital aerial photography as support for restoration, management and habitat monitoring programmes 23 Figure 6 – Pancratium maritimum univariate O-ring statistic. Figure 7 – Bivariate O-ring statistic with Euphorbia paralias as focal species. Figure 8 – Bivariate O-ring statistic with Pancratium maritimum as focal species. FCUP The use of digital aerial photography as support for restoration, management and habitat monitoring programmes 24  Discussion of results Euphorbia plants shows significant intraspecific aggregation when considering radius values up to 3 metres (Figure 5), since the O-ring function curve is greater than upper curve, generated using the confidence envelopes. Pancratium plants shows an even finer aggregation, as it is only significant for radius values up to 2 metres (Figure 6). There was a large area near the centre where none of the species occurred and it represented a large part of the area studied, which could explain the fine aggregation pattern found. According to Figure 7, Pancratium plants show significant segregation from Euphorbia plants for radius values of 2 and 3 metres, since the curve for the O-ring function found is lower than the bottom curve generated using the confidence envelopes. However, for radius values of 18, 23, 24 and 25 metres there is significant aggregation. The opposite situation (Figure 8) shows no significant value for both segregation and aggregation. Looking at Figure 4, we can see that there are two areas of occurrence interrupted by the aforementioned area where none of the species occurs: closer to the shoreline (towards west) and the interior-most area (to the east). This suggests that the presence of Euphorbia could inhibit the presence of Pancratium in some way, perhaps through competition for scarce resources, which would explain the pattern found on the west side. The pattern found on the east side could be explained due to that area being more prone to trampling, which would alter the balance of competition between the two species. The fact that there were no significant values when Pancratium was considered as focal species could possibly mean that Euphorbia has a more dominant role in their interaction. FCUP The use of digital aerial photography as support for restoration, management and habitat monitoring programmes 25 3.2. Management of priority habitat types in mountain areas  Context and objectives One test area was selected from the study areas of a currently ongoing project named LIFE+ Higro (LIFE09 NAT/PT/000043), whose targets are the 4020* and 6230* priority habitats, listed in Annex I of the EU Habitats Directive (see Appendix 1 for a complete list of all habitats referred in the text). This project aims to implement active management plans for the conservation of these habitats, given their poor conservation status in the selected study areas and the threat posed by abandonment of pastoral activities in mountain areas. Applying the techniques developed in this work, the kite system was used to map and analyse the existing habitats and will serve to do so periodically, so that the efficiency of the management measures taken here, as a consequence of the LIFE+ Higro project, can be evaluated.  Study area Serra de Arga is located in the north-western part of Portugal, near Caminha and Viana do Castelo, and inside the Atlantic biogeographic region of Europe. The priority habitats found in the five sections considered are the 4020pt2* and 6230* habitats, as well as other non-priority habitats of interest to the project such as 3130pt2, 4030pt2 and 7140pt2. Both the 4020pt2* and the 6230* habitats can be found in all sectors of the study area, but while the 4020pt2* is well conserved in some places, the 6230* habitat is degraded or very degraded whenever it is found, most likely due to overgrazing and fire. As for the non-priority habitats 3130pt and 7140pt2, they are rare, since they are only found in one section, and very degraded. The 4030pt2 habitat occurs in some areas where the 4020pt2* habitat would occur, but the later was so badly degraded that it was substituted by the 4030pt2 habitat. The RELAPE (threatened or endangered rare, endemic or localized) plant species found were Gentiana pneumonanthe, Genista berberidea, Serratula tinctoria subsp. seoanei and Arnica montana subsp. atlantica. Field work from which this data were obtained was carried out on the 19th of October, 2011. FCUP The use of digital aerial photography as support for restoration, management and habitat monitoring programmes 32 Figure 14 – Zoomed layout of the area chosen for habitat mapping FCUP The use of digital aerial photography as support for restoration, management and habitat monitoring programmes 33 Figure 15 – Grid generated for reclassification. FCUP The use of digital aerial photography as support for restoration, management and habitat monitoring programmes 34 Figure 16 – Reclassification of the habitat mapping. FCUP The use of digital aerial photography as support for restoration, management and habitat monitoring programmes 35  Discussion of results The orthophotomap was visually interpreted by a person with no previous photointerpretation experience, which could lead to some error. Having an experienced technician could improve the accuracy of the results. Moreover, the orthophotomap was generated from photos acquired during a day with some fog, cloudy skies and inconstant wind, meaning that the model obtained could have better quality and could have helped more with the photointerpretation process. Using computer-assisted image classification could also be another technique alternative or complimentary to the one used, especially if the camera used had a multi-spectral lens that could capture the infrared band. In spite of this, the results obtained were considered good and served their purpose quite well, as the different textures of the habitats can be easily identified both in the DTM and the orthophotomap, and the resolution is sufficient to distinguish habitats with similar textures. Also, since the multispectral images obtained from the WorldView-2 commercial satellite have a pixel resolution of approximately 2 metres (http://www.landinfo.com/WorldView2.htm, visited on September 24th, 2012), by creating the 2 metre pixel grid over the image and reclassifying it, a direct comparison between the images can be made, which further augments the possibilities for other types of studies. FCUP The use of digital aerial photography as support for restoration, management and habitat monitoring programmes 36 3.3. Restoration of endangered habitat mosaics in mountain areas  Context and objectives This study area was the target of an ecological restoration programme, an action that was part of a project called ―Valorização e divulgação de habitats naturais nas serras e rios do Baixo Tâmega‖ promoted by the Associação de Municípios do Baixo Tâmega (AMBT). After the diagnosis, a restoration programme was planned and the area must be monitored to evaluate the effect of the measures proposed. The kite system presented in this work served as a basis for the diagnosis and mapping of the habitats and will serve as a tool to monitor the site periodically.  Study area and field surveys The area was named "Turfeira da Almofrela" since it was a peatland area in Serra da Aboboreira (which is part of the Alvão-Marão mountain range), near Almofrela village, in the Baião municipality, Oporto district. This study area was much smaller than the previously described one, (0,5 hectares). However, generally the same priority habitats were found, with the 4020pt2* habitat surrounding the core area, in mosaic with the 4030 habitat, where the 6230* habitat existed in mosaic with the 7140 and 7150 habitats, in spite of its rarity. Other habitats were found, often in mosaic with the aforementioned ones, for example the 3260, 6410 and 6510. This study area also included a patch of ploughed soil that interrupts the continuity of these habitats. The surrounding area showed signs of fire, with some of the 4020pt2* and the 4030 habitats being charred or recovering from charring, while the core area was relatively well conserved but showed signs of eutrophication and was always at risk due to the agricultural activities present nearby. Field work from which this data was obtained was carried out on May 17th and 31st, 2011. FCUP The use of digital aerial photography as support for restoration, management and habitat monitoring programmes 37  Raw DTM's and orthophotomaps Figures 17 to 19 illustrate the main results obtained during the field survey (orthophotomaps and a digital terrain model. Many important aspects of peatland ecosystems and particularly of their vegetation and disturbances can be easily observed. Figure 17 – Orthophotomap of the Turfeira da Almofrela study area. FCUP The use of digital aerial photography as support for restoration, management and habitat monitoring programmes 38 Figure 18 – Zoomed region of the raw orthophotomap of the Turfeira da Almofrela study area. FCUP The use of digital aerial photography as support for restoration, management and habitat monitoring programmes 39 Figure 19 – DTM of the Turfeira da Almofrela study area. FCUP The use of digital aerial photography as support for restoration, management and habitat monitoring programmes 40  Restoration plan proposal Orthophotomaps and the digital terrain model allowed the identification of the areas requiring some type of restoration as well as the best locations for specific actions (Figures 17 to 19). After the diagnosis, the measures needed in order to help the ecosystem recover to a healthier state were chosen. Then, by means of visual assessment of the models generated, the best places to apply the measures were chosen and marked on a map generated from the topographic study of the area (Figure 20). Figure 20 – Proposed restoration plan. FCUP The use of digital aerial photography as support for restoration, management and habitat monitoring programmes 41  Discussion of results The resulting models allowed for a broader view of the area, and made it easier to select the best areas to apply the various measures chosen (Figure 20). Trees were picked to block the wind, to reduce evaporation and were chosen to be placed on the higher parts of the slopes; the peatland area was limited by a fence, in order to avoid stomping, grazing and the alteration of the water surface dynamics and allow the vegetation to regenerate; small obstacles were placed in the water stream that forms after the peatland in order to slow the water flow, reducing erosion and allowing the flooding of the soil, which is necessary in a peatland. Moreover, the quality of the orthophotomaps and of the digital terrain models, together with the cost-efficiency of the methodology, will allow an easy and low-cost monitoring design to be implemented, thereby supporting the assessment of the effectiveness of interventions as well as the possible identification of new ones being necessary. FCUP The use of digital aerial photography as support for restoration, management and habitat monitoring programmes 48 range photogrammetric technique for monitoring slope displacements. 11th FIG Symposium on Deformation Measurements, Santorini, Greece. Sullivan, CA, Bourke, D, Skeffington, MS, Finn, JA, Green, S, Kelly, S and Gormally, MJ (2011) Modelling semi-natural habitat area on lowland farms in western Ireland. Biological Conservation. 144 (3): 1089-1099. Triggs, B, McLauchlan, P, Hartley, R and Fitzgibbon, A (1999). Bundle Adjustment — A Modern Synthesis. ICCV '99: Proceedings of the Internation Workshop on Vision Algorithms, Springer-Verlag. Vanden Borre, J, Paelinckx, D, Mücher, CA, Kooistra, L, Haest, B, De Blust, G and Schmidt, AM (2011) Integrating remote sensing in Natura 2000 habitat monitoring: Prospects on the way forward. Journal for Nature Conservation. 19 (2): 116-125. Walpole, M, Almond, REA, Besançon, C, Butchart, SHM, Campbell-Lendrum, D, Carr, GM, Collen, B, Collette, L, Davidson, NC, Dulloo, E, Fazel, AM, Galloway, JN, Gill, M, Goverse, T, Hockings, M, Leaman, DJ, Morgan, DHW, Revenga, C, Rickwood, CJ, Schutyser, F, Simons, S, Stattersfield, AJ, Tyrrell, TD, Vié, J and Zimsky, M (2009) Tracking Progress Toward the 2010 Biodiversity Target and Beyond. Science. 325 2. Wiegand, T and Moloney, K (2004) Rings, circles, and null-models for point pattern analysis in ecology. Oikos. 104 21. FCUP The use of digital aerial photography as support for restoration, management and habitat monitoring programmes 49 Appendix 1 - List of mentioned habitats This section contains the code and respective designation of all habitats mentioned in the main text. The EU Habitats Directive can be found in the following location: http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=CONSLEG:1992L0043:20070101:EN:PDF 3130 - Oligotrophic to mesotrophic standing waters with vegetation of the Littorelletea uniflorae and/or of the Isoëto-Nanojuncetea 3260 - Water courses of plain to montane levels with the Ranunculion fluitantis and CallitrichoBatrachion vegetation 4020* - Temperate Atlantic wet heaths with Erica ciliaris and Erica tetralix 4030 - European dry heaths 6230* - Species-rich Nardus grasslands, on silicious substrates in mountain areas (and submountain areas in Continental Europe) 6410 - Molinia meadows on calcareous, peaty or clayey-silt-laden soils (Molinion caeruleae) 6510 - Lowland hay meadows (Alopecurus pratensis, Sanguisorba officinalis) 7140 - Transition mires and quaking bogs 7150 - Depressions on peat substrates of the Rhynchosporion