Erich BUHMANN, James PALMER, Matthias PIETSCH and Sarita MAHADIK

Managing the Visual Resource of the Mediterranean Island of Gozo, Malta for Tourists A Studio Approach for International Conversion Students, Bridging Different Levels of English Erich BUHMANN, James PALMER, Matthias PIETSCH and Sarita MAHADIK 1 Introduction The small island of Gozo in Malta is a tourist destination for individuals interested in the Mediterranean Landscape and the rich heritage of Malta. It is assumed that tourists visiting Gozo expect to find a traditional Mediterranean landscape. The Gozitans see the need for economic enterprises and hence they want to develop economic values rather than ecological ones. One purpose of the Landscape Architecture studio was to consider how managing the visual resource in response to tourist expectations and desires might contribute to balancing the need for economic development while protecting landscape ecologic values. The planning agency MEPA has already undertaken the work of mapping and assessing landscape character and sensitivity (MEPA 2004). This map had been developed based on expert interpretation of landscape character. Further visual assessment was conducted by students in the studio, using methods to investigate tourists landscape appreciation, which then become the basis for landscape design guidelines. The studio was composed of students in an International Conversion Program, mostly with very week English language skills. One reason the studio focused on visual landscape resources was that it enabled the use of visual and digital methods that could effectively bridge the different language levels and cultural backgrounds of the students. 2 Material and Methods The purpose of the studio was to develop a concept and methodology to identify landscape classes and formulate design guidelines for each class. The methods employed in the studio include, desk study, field assessment, photographic documentation, informal interviews and a landscape perception study. 2.1 Desk Study and Field Inventory Desk study involved the reading of relevant background reports, mapped information, research works and other data before doing the actual field work. In addition, the students attended a three-day International Conference on Gozo, Malta where eminent scholars and politicians presented research findings about ecological landscape modeling and

Managing the Visual Resource Mediterranean Island of Gozo, Malta for Tourists 65 management for Mediterranean islands. This assisted the students in developing a deeper appreciation for the cultural and ecological issues facing Gozo and other Mediterranean landscapes. During this introductory period of the studio, it became apparent that communicationg primarily through written or spoken English was going to be a barrier to the students effectively working together. Therefore an approach to landscape documentation and analysis that was more visual and graphic was employed. The students spent five days doing fieldwork on Gozo using two distinct approaches for sampling the visual opportunities that tourists were likely to encounter. The first approach involved an analysis of a 10km corridor between the towns of Xlendi and Marsalforn, This corridor crosses the main landscape character types of the island and therefore provides a sufficiently representative transect of the tourist travel experience. A viewpoint was sampled approximately every 500 meters where photographic documentation and a Landscape Assessment Field Sheet were completed. Figure 2 shows the field sheet which inventoried 33 visible landscape attributes. The second approach was to analyze elevated panoramic viewpoints identified in the official tourist map as viewpoint destinations. The 14 most popular viewpoints were chosen for the study. The fieldwork involved completing the Landscape Assessment Field Sheet, along with documenting the full panoramic view and any distinctive character elements at that particular viewpoint. Figure 1 shows the sampled points along the corridor from Xlendi to Marsalforn and the panoramic viewpoints identified on the official tourist map. The transect and the viewpoints cover most of the landscape character areas identified by MEPA (2004) shown in Figure 14.

66 E. Buhmann, J. Palmer, M. Pietsch and S. Mahadik Figure 1. Selected corridor points and viewpoints of tourists destination for study

Managing the Visual Resource Mediterranean Island of Gozo, Malta for Tourists 67 Figure 2. The Landscape Assessment Field Sheet formulated for the study area Field data was collected in a rigorous way through the use of the Landscape Assessment Field Sheet. The field sheet provided a systematic description of landscape character and identified the current landscape condition. Sketches in the form of road sections were used to illustrate the landscape character. Photographs were taken to document the full 360 degrees of the view from every study point and the location of the point was identified using a cell phone GPS. The photographs were organized into four panoramic sections of 90 degrees each, as this seemed a reasonable approximation of a human view. Detailed photos of particular elements were also recorded. During the fieldwork, students took the opportunity to conduct informal interviews with the local citizens to develop a more holistic understanding of their landscape perceptions and opinions, as well as assisting the students to better read the landscape.

68 E. Buhmann, J. Palmer, M. Pietsch and S. Mahadik 2.2 Identify Representative Photos The photos gathered during the fieldwork were evaluated to identify a representative sample for further evaluation. This involved the following steps: Prepare a photo inventory of more than 2000 photos gathered during the fieldwork represents a full range of Gozo landscapes (e.g., field with rubble wall, fields without rubble walls, city playa, city street, city on the horizon, small cluster of development surrounded by fields, etc.). Photos that lacked clarity or were otherwise flawed were excluded from the photo inventory. Develop a landscape classification based on the typical landscape elements that characterize Gozo. Five general types of landscapes were identified: Sea, Fields/Agriculture, Non fields, Settlement and Streetscape. The photos for public evaluation were a set of 40 photos that represented the landscape elements and tourists landscape experience. Of the 40 photos selected for public evaluation, 18 focused on Streetscape while the remaining 22 represented the other classes. 2.3 Landscape Perception Survey A landscape perception survey was conducted using the representative photos to better understand how the Gozo landscape is perceived by tourists. A total of 201 respondents participated in the landscape perception survey. 107 were students from Bangladesh, China, Germany, India, Indonesia, Iran, Thailand, USA and Vietnam. 94 participants lived in Germany, but were not students; they added diversity of age and experience. While not a random sample, these respondents represent the cultural and life experience diversity that might be expected of tourists. A Landscape rating scale was created by setting out five cards on a table representing a scale of scenic quality from (1) Least to (5) Most scenic, with (3) being Average scenic quality, as illustrated in Figure 3. 5 Most 4 3 Average 2 1 Least Figure 3. The grading scale from 5 (Best) to 1 (Least) for the public convenience Participants rated the 40 photos by placing them along the rating scale. The rating for the photos from each pile were then recorded in the evaluation sheet shown in Figure 4. The photo ratings were averaged over all the participants. Based on this analysis, five photos were identified that represent the most scenic, the closest to average, and the least scenic of the landscapes that tourist might commonly see.

Managing the Visual Resource Mediterranean Island of Gozo, Malta for Tourists 69 2.4 Graphic Evaluation of Landscape Character Measuring the components of each landscape scene was done to better understand what contributed to Gozo s most, average, and least scenic quality. The approach used is described by Shafer and Mietz (1970) in their article It seems possible to quantify scenic beauty in photographs. The first step was to carefully separate the foreground from the middle/background with a drawn line. (The middle and background was treated the same, since there is very little background except for the Sea.) Then the foreground and middle/background were divided into basic landscape types (e.g., sea, field, non-field, settlement, streetscapes). Next each the elements that compose each landscape type were identified and their area is measured using a 1cm-by-1cm grid placed over each photo, as shown in Figure 7. The percent of the earth s surface in the view that each element and type within the foreground and middle ground/background occupies is recorded. Figure 8 shows a typical data summary for a typical photograph; Figure 9 gives the data summary for all 40 photographs.

70 E. Buhmann, J. Palmer, M. Pietsch and S. Mahadik Figure 4. Landscape Perception Survey recording sheet and participants taking the survey. Fig. 5: The distribution of ratings for each photo

Managing the Visual Resource Mediterranean Island of Gozo, Malta for Tourists 71 Figure 6. Findings from the landscape photo evaluation survey

72 E. Buhmann, J. Palmer, M. Pietsch and S. Mahadik Figure 7. The 1cm by-1cm grid and the percentage distribution among the foreground, middle ground and background Figure 8. A typical data summary for the landscape types that comprise a view

Managing the Visual Resource Mediterranean Island of Gozo, Malta for Tourists 73 3 Data Analysis The five best, five average, and five least scenic views from the landscape perception survey were then evaluated for their dominant landscape elements. The results guided the students interpretation of what particular landscape features appear to determine Gozo s landscape scenic quality. 3.1 Best Rated Images Figure 10 shows three of the five most scenic views and shows the profile of their landscape types. A more detailed analysis found that the landscape elements that characterize these five views are: sea line, open landscape, rural landscape elements, church on the horizon, and street with trees on both sides. Figure 10. The landscape elements that compose three of the most scenic views. 3.2 Average Rated Images Figure 11 shows three of the five average scenic views and shows the profile of their landscape types. A more detailed analysis found that the landscape elements that characterize these five views are: one row of trees along the driveway, single row of trees, individual agricultural patches, trees in urban edge, and trees in median.

74 E. Buhmann, J. Palmer, M. Pietsch and S. Mahadik Figure 11. The landscape elements that compose three of the average scenic views.

Managing the Visual Resource Mediterranean Island of Gozo, Malta for Tourists 75 3.3 Least Rated Images Figure 12 shows three of the five least scenic views and shows the profile of their landscape types. A more detailed analysis found that the landscape elements that characterize these five views are: roads, car in parking, disturbed land, erosion, and cement wall. Figure 12. The landscape elements that compose three of the least scenic views. 4 Results Landscape character reflects particular combinations of geology, landform, soils, vegetation, land use and human settlement. It creates the particular sense of place experienced in different areas of the landscape. Landscape character is sufficiently generalized that it can be easily mapped, as shown by MAPA (2004). The landscape types which were developed by the studio as reported here also results in a type of landscape visual assessment that can be maped. Landscape elements that compose the landscape types can not be easily map, though they are clearly visible in the landscape.

76 E. Buhmann, J. Palmer, M. Pietsch and S. Mahadik Fig. 13: Erosion is a negative foreground element, but in the background when combined with topographic relief, it was rated high. Fig. 14: Streetscapes without trees are rated negatively while ones with green are rated positively. Further findings of this study For the results for favorable landscape elements by visitors, there is no difference in evaluation between Students and Non students Clear favorites were landscape variables representing more rural and open landscape. Streetscapes with trees were more appreciated The Landscape groups that were more appreciated were Sea, Fields and Non fields The Landscape elements that were more appreciated were waterfront, prickly pear fence, rubble walls, trees, Garigue and Maquis Landscape

Managing the Visual Resource Mediterranean Island of Gozo, Malta for Tourists 77 There were two types of results: results that can be related to spatial land use and results related to single landscape features. The large number of evaluations are valid for statistical analysis which has not yet been done. However, the spatial related were further processed using GIS. A map of favorable landscape elements can be generated with design criteria to improve the landscape visibility based on the results The map can then be compared with the landscape accessibility report. 5 Implementation of Results Before further implementing the results, the findings had been compared with the Landscape Assessment Study by MEPA done in 2004 for GOZO and for Malta. In this paper, only the Gozo parts of the Landscape Assessment study are quoted. The following flow diagram shows the GIS procedure for this combination. Fig. 15: GIS workflow unifying the geographic and the behavioral visual evaluation of the landscape of Gozo. In the Visual Landscape assessment study, the island of Gozo had been divided by MEPA into 35 landscape character units. Classification has been based upon the predominant landscape elements, topography and zones of visual influence. MEPA defined landscape character Units (Figure 14) and based on the character of landscape units they defined sensitivity areas and made suggestions for potentially upgrading the landscape (Figure 16).

78 E. Buhmann, J. Palmer, M. Pietsch and S. Mahadik Fig. 16: Landscape Character Areas. (Source MEPA, 2004) Based on the expert evaluation of these units, a Landscape Sensitivity Map (Figure 15) and Landscape Upgrading potential Map (Figure 16) had been developed Fig. 17: Landscape Sensitivity Map. (Source MEPA, 2004)

Managing the Visual Resource Mediterranean Island of Gozo, Malta for Tourists 79 Fig. 18: Landscape Upgrading Potential Map (source MEPA, 2004) - MEPA further interpreted the landscape potential map. The expert evaluation was now compared with the visual assessment study (user perspective) Figure 17 shows the summary of main landuse by MEPA: It was feasible to translate the evaluation of the spatial landscape features to this landscape evaluation Fig. 19: Landuse Map of MEPA (source MEPA, 2006) - The landuse map of MEPA is also based on landscape categories. Reattributing this evaluation to the five appreciation levels shows the spatial representation of the visual evaluation by visitors (Figure 18)

80 E. Buhmann, J. Palmer, M. Pietsch and S. Mahadik Fig. 20: Landscape Visual quality of Landscape (source MEPA, 2006) Evaluation by Anhalt University 2011 Comparing both maps in Figure 19 shows the close evaluation of the expert geographic approach to the public evaluation. Fig. 21: Landscape Correlation Map, 2011 - shows high correlation between the two approaches

Managing the Visual Resource Mediterranean Island of Gozo, Malta for Tourists 81 Presenting these methods to Gozitans may be a second catalyst, as the discussion of landscape character and scenic quality from a tourist s perspective may suggest new opportunities for Malta s government officials, planning staff and citizens. Fig. 22: Example showing a design approach for the corridor points.

82 E. Buhmann, J. Palmer, M. Pietsch and S. Mahadik Fig. 23: Example 01- shows a design approach for the View points.

Managing the Visual Resource Mediterranean Island of Gozo, Malta for Tourists 83 Fig. 24: Example 02- shows a design approach for the view points. 6 Conclusion The study shows that applying methods of a visual landscape inventory and a landscape perception study can be analized to develop design guidelines. based on the results of the analysis work as a studio approach for international conversion students, bridging different levels of English. The use of predominately visual methods could be a catalyst for teaching basic research methods to students with different cultural backgrounds and who have different skill levels in academic reading and writing in English. Dealing with visual aspects such as field work, evaluation, analysis and design of suggestions addresses the design skills which are often very well developed and not limited by the level of knowledge of the teaching language. This paper documents the evaluation of the survey and the comparison with the geographic approach by MEPA and some recommendations based on favorable landscape elements and composition. The University of Malta will now interview Gozitans using the same set of photos. We are confident that the results will be comparable. Further analysis and translation into implementation on Gozo will be discussed by the local authorities and the public. 7 References BISHOP. I. HULSEB,D.W. ( 1994 ) : Prediction of scenic beauty using mapped data and geographic information systems In: Landscape and Urban Planning 30 ( 1994) 59-70

84 E. Buhmann, J. Palmer, M. Pietsch and S. Mahadik JESSEL, B. FISCHER-HÜFTLE,P., JENNY,D. & ZSCHALICH, A. (2003): Erarbeitung von Ausgleichs- und Ersatzmaßnahmen für die Beeinträchtigung des Landschaftsbildes (Schriftenreihe Angewandte Landschaftsökologie 53) Bonn Bad Godesberg MEPA Malta Environment & Planning Authority (2004): Landscape Assessment Study http://www.mepa.org.mt/file.aspx?f=4180 (accessed November 11, 2010). NOHL,W. (2009): Grünland und Landschaftsästhetik. Die ästhetische Bedeutung von Grünland und die Auswirkungen vermehrten Grünlandumbruchs auf das Landschaftsbild. In: Naturschutz und Landschaftsplanung, 41 (12) S. 357-364 PALMER, J.F., ROOS-KLEIN LANKHORST, J, (1998): Evaluating visible spatial diversity in the landscape. In: Landscape and Urban Planning 43 (1998) 65-78 SHAFER Jr., ELWOOD L., and MIETZ, J. (1970): It Seems Possible to Quantify Scenic Beauty in Photographs, Research Report NE-162. Upper Darby, PA: USDA, Forest Service, Northeastern Forest Experiment Station. SWANWICK, C. and LAND USE CONSULTANTS (2002): Landscape Character Assessment: Guidance for England and Scotland: The Countryside Agency, & Scottish Natural Heritage THE LANDSCAPE INSTITUTE AND INSTITUTE OF ENVIRONMENTAL ASSESSMENT (1995): Guidelines for Landscape and Visual Impact Assessment. London: E & FN Spon. energy consumption WÖBSE, H.H (2002): Landschaftsästhetik, Stuttgart