Towards objective evaluation of UNESCO Global Geoparks using a
Geodiversity Index, in Fiordland National Park, New Zealand
Roos van Wees -10736441 03-08-2017, Amsterdam
Supervisor: dhr. dr. A.C. Seijmonsbergen Second supervisor: dhr. dr. W. M. de Boer
Front page image is an example of a Geodiveristy Index (with equation 2, cell size 750m) of Fiordland National Park The left logo is of the University of Amsterdam and the right logo is of UNESCO Global Geoparks (2017)
Abstract
Geoparks are parks with important geological heritage which are established by UNESCO all over the world to promote protection, education and sustainability of geodiversity. Abiotic environmental factors such as geology, soils and landforms are often considered less vulnerable and therefore less valuable. Destruction of abiotic resources is often irreversible. Geodiversity has intrinsic, cultural, aesthetic, economic, educational and scientific value, nevertheless objective evaluation of geodiversity is still an open issue. To value geodiversity, information about independent criteria is lacking and there is a need to study these factors in different spatial context. Geodiversity can be quantified by using a Geodiversity Index (GDI) of different abiotic factors. A GDI is made in ArcGIS 10.4 of Fiordland National park in New Zealand with different grid sizes of 500, 750 and 1000 meter and four GDI formulas. Six sub-indices are created to calculate the GDI’s: Geology index, Geomorphology index, Soil index, Hydrology index, Elevation index and a Slope index. The geodiversity in the GDI’s where divided in 5 classes from very low, low, medium, high and Very high diversity. After evaluation of the variance the most appropriate cell size for Fiordland National park turned out to be 750 meter with equation 2 (Geology + Geomorphology + Soil + Hydrology + Elevation + slope*0.5). The resulting geodiversity map is interpreted in terms of clusters of High and Very high geodiversity patterns, and compared to the original patterns of the input data. It is recommended that a GDI should be developed for a potential Global Geopark to enable objective evaluation of geo-factors by experts of UNESCO. This information can be used when going into the field of the applicant Geopark.
Table of content
1. Introduction
1.1 Theoretical framework………5
1.2 UNESCO Global Geoparks background………..5
1.3 Study area……….6
1.4 Aim……….7
1.5 Research questions………7
2.
Methods
2.1 Workflow and Overview……….82.2 Data collection……….8
2.3 Pre-processing; creating the sub-indices………8
2.4 Calculating sub-indices………8
2.5 Calculation of the GDI……….9
3. Results
3.1 Visualization and analysis of the sub-indices and GDI maps………114. Discussion
4.1 Interpretation of the results……….……….144.1.1 Different GDI formulas 4.1.2 Geodiversity clusters and patterns 4.2 Recommendations for objective evaluation of Geoparks……….15
4.3 Limitations and Further research……….16
5
.Conclusion………...16
References………..…….17
Appendix A, B, C and D
1.
Introduction
1.1 Theoretical framework
Throughout the years, the understanding of Earth’s diversity and dynamics led to several new concepts and terms. One of these terms is geodiversity. Cendrero (1996) was the first to state that the intrinsic value of geological elements should be considered when classifying a geological area of interest. In further research Gray (2004) defined geodiversity as: “Geodiversity: the natural range (diversity) of geological (rocks, minerals, fossils), geomorphological (land form, processes) and soil features. It includes their assemblages, relationships, properties, interpretations and systems”. This was followed upon by Kozlowski (2004) who included surface waters.
The value of geodiversity is not to be underestimated, it is dynamic, transforms, uses and produces energy even without the interference of biology (Parks and Mulligan, 2010). Gray (2004) valued geodiversity in four groups: the intrinsic value, the cultural and aesthetic value, the economic aspect of resources and the scientific and educational value. Moreover, valuating geodiversity will bring a better view on the natural resources which will help to manage them and support a sustainable use (Serrano and Ruiz-Flaño, 2007; Panizza, 2008; Ruban, 2010; UN, 2016). In addition, it will promote awareness on geological hazards and helps prepare for disaster mitigations strategies in these areas (Giardino et al., 2012; UN, 2016). Further on geodiversity in national parks will hold record of past climate change and are evidence and educators of the effect of current climate change (Crozier, 2010; UN, 2016). Finally,
geodiversity can spread awareness of geological heritage and all the links it contains to natural, cultural and intangible heritage (UN, 2016). Geodiversity is considered less vulnerable than biodiversity and cultural heritages which results in less (scientific) attention (Reynard and Coratza, 2007). Nevertheless, if features of geodiversity are damaged or disturbed, it is irreversible or takes whole geological time periods to recover (Prosser et al., 2010). All these values emphasize the need for protection of geodiversity of an area.
The concept of geoconservation is connected to geodiversity and is defined as “the active management of landscapes to conserve and enhance geological and geomorphological features, processes, sites and specimens” (Seijmonsbergen et al., in press). An objective evaluation system of the geodiversity in an area of interest, will benefit geoconservation and management. However, valuing of abiotic features is not often based on objective and transparent criteria (Bruschi and Cendrero, 2005). Moreover, there is a lack of information about geodiversity in different spatial context (Serrano and Ruiz-Flaño, 2007; Hjort and Luoto, 2010). To objectively evaluate the geodiversity a Geodiversity Index (GDI) can be calculated for the area of interest, as has been demonstrated in various areas (Serrano and Ruiz-Flaño, 2007; Parks and Mulligan, 2010; Hjort and Luoto, 2010; Melelli, 2014).
1.2 UNESCO Global Geoparks Background
The United Nations recognized the importance of abiotic nature and founded the Global Geoparks
Network (GGN) in 2004, it consists of UNESCO Global Geoparks all over the world, all including geological heritage of international value (UNESCO, 2016). Anno 2017, 127 Geoparks are established in 35 different countries (UNESCO, 2017c) UNESCO promotes education, sustainable development and connection between the Geoparks, moreover they use a bottom-up approach and engage the local community in decisions and management about the concerning Geopark (UNESCO, 2016). The founding idea is that the scientific and social society can benefit from geodiversity and that geological heritage should be
conserved. Objective valuation of the geodiversity in Geoparks can be provided by geodiversity mapping and creating a GDI as has been undertaken before (Benito-Calvo, Perez-Gonzalez, Magri and Meza, 2009; Melelli, 2014). However, in the application process to become a Geopark it is mentioned that, among other steps, experts and professionals will review the Geopark (UNESCO, 2017b).Although subjectivity is an almost unavoidable part of the experts review on the parks (Bruschi and Cendrero, 2005), an
independent evaluation system would encourage evaluation based on objective criteria, to be applied and transferred to other Geoparks. Several attempts exist to create an objective procedure for the evaluation of Geosites, that emphasize that differences in the evaluation of the landscape are difficult to avoid (Bruschi and Cendrero 2005; Reynard, Fontana, Kozlik and Scapozza, 2007). However, these differences stress the need for improved procedures and use of a geodiversity index-based method for a Geosite or Geopark. Until now, the use of a geodiversity index in any evaluation method of UNESCO has not been mentioned. In this study, recommendations on the use of a geodiversity index will be presented for use in the application process for Geoparks.
1.3 Physiography of Fiordland National park
The focus area of this study is Fiordland National park (https://www.fiordland.org.nz/about-fiordland/fiordland-national-park/) which lies in the UNESCO World Heritage Site Te Wāhipounamu (Māori for "the place of greenstone"). Fiordland is the largest national park of all the 14 parks in New Zealand and covers an area of 12500 km2. It is located in the southwest corner of the southern island of New Zealand (see Figure 1). The northern tip is marked by the fault zone of the southern Alpes between the Indo-Australian and Pacific continental plates which makes it one of only three segments of the world’s major plate boundaries on land (Walcott, 1998; UNESCO, 2017a). It mainly consists of crystalline rock (gneiss and granites) and it is a plateau with 1000-meter elevation, this plateau is eroded in the Pleistocene glaciation period which formed deep carved fiords and typical u-shaped valleys (Walcott, 1998). The present-day Fiordland is almost ice free, but still owes a geomorphology to glaciations which retreated 10.000 years ago (Augustinus, 1992; Wing and Jack, 2014). Furthermore, in these valleys Tertiary limestone occurs.
Figure 1. Study area (Google Earth and Google maps, 2017)
Figure 2. Location of inland and coastal fiords, Fiordland, New Zealand
This is suitable for cave forming which tend to yield a lot of fossils as for example the Te Anau caves in the east of the park (Garnock-Jones, Bayly, Lee and Rance, 2000). Moreover, several landslides have occurred in the park, mainly due to slipping of Tertiary mudstone (Hancox and Perrin, 2009). Fiordland undergoes strong seismic activity due to the subduction of the oceanic lithosphere of the Australian plate under Fiordland, resulting in several convergent and transcurrent faults in the park and uplifting peaks (Davey and Euan, 1982; Walcott, 1998). Fiordland consists of many acid brown to peaty acid brown soils and below the treeline the soils are dominated by orthic podzols (Hewitt, 1998). It is a temperate rain forest and receives to > 8m a year precipitation, which is similar to the western southern Alpes (Wing and Jack, 2014). In fig 1. can be seen that the Fiordland has a clear drainage pattern, water flows through the valleys into the fiords seen in fig. 2 (Augustinus, 1992; Google Earth and Google Maps, 2017). The park resembles great similarity to the biota of Gondwanaland and is due to the late settlement and rough terrain largely unmodified with indigenous species (Davey and Euan, 1982; Walcott, 1998; Wing and Jack, 2014).
1.4 Aim
As previously mentioned, there is need to study geodiversity with objective criteria and fill knowledge gaps in different spatial environments (Bruschi and Cendrero, 2005; Hjort and Luoto, 2010). Calculating a GDI for Fiordland National Park will partly fill the knowledge gap of spatial context in this latitude and longitude, considering this will be the first GDI of an area in New Zealand. Therefor it will evaluate the geo-factors and has potential as a conservation tool, necessary to investigate long term conservation of abiotic factors (Anderson and Ferree, 2010; Parks and Mulligan, 2010; Crozier, 2010; UNESCO, 2016). The overall aim of this study is to investigate how geodiversity can be used to objectively evaluate abiotic nature in natural protected areas, with a case study on Fiordland National Park. The aim is split into: (1) to make an inventory of existing abiotic elements and to develop a GDI of Fiordland National Park in New Zealand; (2) To give recommendations about how to use the GDI in the guidelines of UNESCO for establishing a Geopark.
1.5 Research questions
The following research questions should be answered to reach the aims: How can a geodiversity index be used to enhance preservation of geodiversity in natural protected areas, a case study on national Park Fiordland of New Zealand?
- What is the most appropriate cell size for the construction of a GDI in the light of the input factors? - What formula for the GDI should be applied to Fiordland National Park?
- How should the geodiversity map be interpreted in terms of geodiversity patterns?
2. Methods
2.1 Workflow and Overview
This study uses Geographic Information Systems (GIS) as it is a widely used analytical tool for
quantitative analysis of spatial context, relationships between geo-factors and to create numerical indices (Melelli, 2014). The GDI is a compound index that includes the geological diversity, geomorphology diversity, soil diversity, drainage diversity, elevation diversity and slope diversity, which are added as sub-indices. Every GDI will be created using a 500, 750 and 1000 meter grid size and there will be
experimented with different GDI formulas to eventually choose the best grid size and formula. The general workflow is presented in fig. 3, all steps will be addressed in detail in separate paragraphs.
2.2 Data Collection
The collection of different data of a certain area gives a database with differences in factors as legend units, quality, coverage and age (Seijmonsbergen et al., in press). Most of the data is national data for New Zealand or specific for the south or north island. The geological, geomorphological, DEM and soil data are all retrieved from the Land Resource Information System Portal (LRIS Portal) (see table 1). In addition, the hydrology and boundary of Fiordland are derived from the Land Information New Zealand Data Service (LINZ) (see table 1). The topographical (elevation and slope) information is derived from a 25 meter resolution DEM, the slope map was made with the slope tool in the spatial analyst. The hydrology map was created from two datasets of available rivers and lakes in ArcGIS. Every branch of a river has a different value (see table 1).
The boundary shape of Fiordland was collected from a dataset of protected areas in New Zealand. The land surface parameters as elevation and slope are considered part of the geodiversity, in this study they are not equal to geomorphology, but to see the influence of these two parameters the GDI formulas will be adjusted (Räsänen et al., 2016). In table 1 the used data listed. All the data was set to the NZGD2000 / New Zealand Transverse Mercator 2000, coordinate system in ArcGIS. The metadata and information about the content of the data is found in table 1.
Figure 3. Workflow of this study 1. Data collection 2. Data pre-processing 3. Calculating the Geodiversity Index 4. Grid definition 5. Visualisation of the sub-indices and the GDI
6. Detecting 'hotspots’ and patterns of geodiversity 7. Recommendations for Geoparks
2.3 Data pre-processing; creating the sub-indices
The different sub-indices for the calculations of the GDI are used, which have all been classified into five classes ranging from Very low, Low, Moderate, High to Very high, following the method used by Kozlowki (2004). The hydrology map contained different types: polylines (rivers) and polygons (lakes), which cannot be merged. At first, the polylines where converted to a raster (polyline-to-raster tool) and next polygons where made with the raster to polygon tool, so that rivers and lakes where both polygons and could be merged. In the data of the National Protected areas, Fiordland National Park was selected and then exported as a separate layer for intersecting the sub-indices with the intersect tool. After clipping all maps to the extent of the Fiordland boundary, the geology, geomorphology, soil and hydrology polygon maps where converted to raster maps with a 25 meter cell size, similar to the DEM. The ‘Create-fishnet’ tool was used to generate three fishnets of 500, 750 and 1000 meter. Then, the number of different units where counted per fishnet grid with the ‘Variety’ option in the ‘Zonal Statistic’ tool. For the ‘Elevation’ and ‘Slope’ rasters the ‘standard deviation’ option for each fishnet size is used, because the ‘variety’ option does not correctly present the diversity of the slope angles (Bray, 2016). All the rasters contained more than 5 units per cell (see table 2), so the Reclassify tool was used to create the five classes with the ‘natural break’ method. The ‘natural break’ method is used because it maximized the variance between classes. This will result in six sub-indices: (1) Geological sub-index (Gdi), (2) Geomorphological index (Gmdi), (3) Soil index (Sdi), (4) Hydrology index (Hdi), (5) Elevation index (Edi) and (6) a Slope index (Sldi).
Sub-index Contains Scale /
cell size Type Feature count No. Values / range
Data set name / source / year
Geology Geological index
Geological units 1:250 000 Vector multipolygo n
1321 12 South Island Soilscapes, 2011
https://lris.scinfo.org.nz/layer/126-south-island-soilscapes/ Geomor – phology Geomorpholog ical index Geomorphologic al units
1:50 000 Polygons 497833 25 LCDB v4.1 - Land Cover Database version 4.1, Mainland New Zealand, 2015
https://lris.scinfo.org.nz/layer/423-lcdb-v41-land-cover-database- version-41-mainland-new-zealand/ Soil
Soil index Soil classes 1:50 000 Polygons 107298 32 FSL New Zealand Soil Classification 2010 https://lris.scinfo.org.nz/layer/126-south-sland-soilscapes/metadata/ Surface waters Hydrology index Rivers Lakes 1:50 000 Polylines Polygons 3 22331 (river branches) 3120 (lakes)
NZ Lake Polygons (Topo, 1:50k), 2010 https://data.linz.govt.nz/layer/293-nz-lake-polygons-topo-150k/
NZ River Centrelines (Topo, 1:50k) , 2010 https://data.linz.govt.nz/layer/327-nz-river-centrelines-topo-150k/ DEM (Digital Elevation Map) DEM (Digital Elevation Map) - Elevation - Slope Topography: - Slope, - Elevation 25m 1:50 000 Raster - -0,8248 – 2722,26
Landcare Data Use License, 2016
https://lris.scinfo.org.nz/layer/127-nzdem-south-island-25-metre/ Boundary Fiordland - Boundary of the Fiordland
1:50 000 Polygon 17649 1 Protected areas 2017
https://data.linz.govt.nz/layer/3564-protected-areas/
500 meter grid size 750 meter grid size 1000 meter grid size Geology 5 5 5 Geomorphology 8 7 9 Soil 5 5 6 Hydrology 18 29 27 Elevation 0-306.2 0-366.4 0-422.0 Slope 0-28.7 13-18.1 0-27.6
2.4 Calculation of the GDI
The Raster Calculator tool was used to create four GDI’s for 500, 750 and 1000 meter cell size, with as result 12 maps in total. The following equations will be used and evaluated in the discussion:
1. GDI = Gdi + Gmdi + Sldi+ Hdi + Edi + Sdi 2. GDI = Gdi + Gmdi + Sldi+ Hdi + Edi + Sdi * (0.5) 3. GDI = Gdi + Gmdi + Sldi+ Hdi + Edi
4. GDI = Gdi + Gmdi + Sldi+ Hdi
All sub-indices are added for every grid size (500, 750 and 1000m). This is then categorized into five classes with the ‘slice tool’ ranging from Very low, Low, Moderate, High to Very High. After creation of the sub-indices and the GDI’s statistics are calculated with the Zonal Statistics as Table function and a
correlation matrix is prepared with the Band Collection Statistics. This is done to show the correlation between the sub-indices and the GDI’s to evaluate the influence of every individual sub-index on the GDI, all correlation matrices can be seen in Appendix D.
3. Results
3.1 Visualization and analysis of the sub-indices and GDI maps
In the results, the GDI’s of equation 2 and 4 are illustrated, the sub-indices and the GDI’s of equation 1 and 3 are presented in Appendix A and B, this to show the difference between the GDI’s with the highest and lowest variation in the results (see table 4). In addition, statistical analyses are represented and deliberated on.
The classes of High and Very high geodiversity occur in the regions with a complex topography as can be seen in the slope map, figure 4. Visual inspection of the slope map shows that the slope seems to reflect
the patterns present in the various GDI’s, see figure 4, 5 and 6. The difference of equation 2 and 4 is especially visible in the high geodiversity part in the north of Fiordland National Park (see blue circles in figure 5 and 6). There can be seen that equation 2, contains more High and Very high geodiversity in this area than equation 4. When comparing figure 5 and 6 it is clearly that scale size 750 and 1000 meter consistently contain more Very low (dark green) cells compared to the 500 meter cell size.
There are differences in the standard deviation and highest diversity class for every equation and fishnet grid. Equation 3 (without slope) was the only equation of which the ‘Low’ diversity class was the most represented in the GDI for every scale, with a range of 29% to 35% seen in table 3. Equation 4 with only four sub-indices showed the highest diversity in the lowest two classes (Very low and Low) and had the lowest standard deviation ranging from 1.0 to 1.04, see table 4. Therefore, equation 1 and 2, with scale size of 750 meter have the highest standard deviations of 1.17 and 1.26 (see table 4).
Table 3. Standard deviations and highest diversity classes of all cell sizes and equation
Figure 5. GDI’s of equation 2 of 500, 750 and 1000 meter grid size Figure 4. Unprocessed slope map,
The chosen fishnet grid depends on the cell size, quality and input data (Seijmonsbergen et al., in press). Hengl (2006) recommends the grid resolution that is in between the Coarsest and the Finest legible grid resolution. In which the coarsest can be used to show complexity of the terrain and rough hotspots while the finest presents 95% of all spatial objects or topography, but increases computation time (Hengl, 2006).
The geodiversity map with the most variation is the most representative for the geodiversity. This means that out of the four geodiversity scenarios times three different grid sizes the scenario with the highest standard deviation reflects the most appropriate grid size. The highest standard deviation of 1.26273 is related to the 750 grid size of equation 2 (see table 4). This is in line with a former bachelor thesis of Bray (2016), in which is stated that 750 meter is the best grid size for conservation of National Parks.
500 meter grid size 750 meter grid size 1000 meter grid size
Equation 1: Very Low 12% 19% 16%
Low 34% 29% 29%
Medium 25% 28% 31%
High 23% 17% 18%
Very high 6% 7% 7%
Equation 2: Very Low 19% 26% 20%
Low 30% 29% 31%
Medium 30% 19% 29%
High 18% 17% 14%
Very high 2% 8% 5%
Equation 3: Very Low 27% 26% 27%
Low 35% 29% 29%
Medium 27% 28% 25%
High 9% 13% 16%
Very high 2% 4% 3%
Equation 4: Very Low 21% 23% 21%
Low 37% 32% 37%
Medium 31% 32% 31%
High 9% 11% 9%
Very high 2% 3% 2%
Figure 6. GDI’s of equation 4 of 500, 750 and 1000 meter grid size
Table 5 presents the correlation matrix of all the sub-indices and the GDI of equation 2 with 750 cell size. The highest correlation is between the slope and the GDI, which is 0.93075. Moreover, the
Geomorphology and GDI have a high correlation of 0.75156. The highest correlation between two sub-indices are slope and geomorphology, with a correlation of 0.72753, see table 5. Moreover, the correlation between the sub-indices Geology and Elevation is negative, so when an area has a high variance in geology than there is a low elevation diversity and vice versa.
500 meter cell size 750 meter cell size 1000 meter cell size
Equation 1 1.102929 1.17046 1.126445
Equation 2 1.0856 1.26273 1.12227
Equation 3 1.00071 1.10708 1.13148
Equation 4 0.9583 1.0382 1.0113
CORRELATION
MATRIX GDI 750m Soil Geomorphology Elevation Hydrology Slope Geology
GDI 750m 1 Soil 0,65932 1 Geomorphology 0,75156 0,36383 1 Elevation 0,49929 0,12875 0,36049 1 Hydrology 0,41270 0,12887 0,12543 0,0173 1 Slope 0,93075 0,60422 0,72753 0,57109 0,40094 1 Geology 0,48676 0,62853 0,14642 -0,04249 0,10921 0,41946 1
Table 4. Standard deviations of all grid sizes and equations
4. Discussion
4.1 Interpretation of the results 4.1.1 Different GDI formulas
The equations 1 and 2 have the highest variance, which results in a higher standard deviation. This can be explained because equation 3 and 4 have less sub-indices which is less unit per cell and this will ultimately result in a lower standard deviation compared to equation 1 and 2.
According to Tabachnick and Fidell (1996) a correlation between two independent variables > 0.7, suggests there is multicollinearity present in this model. One way to correct the collinearity would be to drop one of the sub-indices, according to this the slope is dropped in equation 3 and 4. In the correlation matrix of equation 3 and 4 there is no correlation > 0.7, so no multicollinearity, see Appendix D.
The correlation matrix of cell size 750 of equation 3 and 4 does not show any collinearity, due to the dropped index of slope. This does result in a higher correlation of the Soil and Geomorphology sub-index with the GDI of equation 3 and 4 (see Appendix D).
4.1.2 Geodiversity clusters and patterns
The selected GDI consists of different clusters and patterns of geodiversity. Interpretation of patterns can only be considered with the original input indices in mind, these can be seen in Appendix C (Seijmonsbergen et al., in press). In figure 7 can be seen that roughly 3 clusters of High and Very high geodiversity can be distinguished. The blue circle in the northern part of the Fiordland is the most southern part of the Southern Alps and Alpine fault which stretches across the southern island of New Zealand (University of Otago, n.d.). This part is characterized by rapid elevation changes which create a steep slope diversity as well. This area is popular amongst the tourists, because of the Milford Sound, which is the only fiord accessible by road in New Zealand, it lies in the middle of this blue circle (see white cells which indicate the fiord) (DOC, 2017). The Milford sound is the wettest place in New Zealand with an annual rainfall of 6412 mm (Wing and Jack, 2014).
In the center, the Very high diversity (value 5) is less frequently present. Nevertheless, the occurrence of red clusters is mainly explicable by the high diversity of slopes and variation of geomorphological units of in this region, see Appendix B.
Lower lying areas consist of more geology compared to the higher elevation areas of Fiordland, this is confirmed by a Figure 7. GDI (equation 2) with highest variation,
negative correlation in the correlation matrix (table 5 and see Appendix C for the input Geology map). The geology is more complex along the rivers and in the south coast, here are more sedimentary rocks, most likely the effects of sedimentation by rivers.
The southern part of Fiordland National Park has several clusters of higher diversity. According to the sub-indices the slope has the highest correlation and this area is part of a mountain range. This causes a high influence of the elevation and slope sub-indices. Also this area is known amongst tourists, because of popular Great Walks which are situated in this mountain region near the city Te Anau (Department of Conservation, n.d.). Moreover, in the lower elevation parts the soil sub-index has more diversity, see Soil map and Index in Appendix B and C.
4.2 Recommendations for objective evaluation of Geoparks
The geodiversity index has potential to be used as evaluation tool for a National Park and Geoparks. If all Geoparks are mapped using a GDI, the most important clusters and high diversity patterns can be rapidly evaluated. This can support the park management and raise its sustainability (Melelli, 2014). Such knowledge can be exchanged and improve connections between Geoparks in the Global Geoparks Network (GGN).
The application process to become an UNESCO Global Geopark is found on the UNESCO website and in reports (UNESCO, 2015; UNESCO, 2017b; UNESCO, 2017c). A short summary of the application procedure has four steps:
1. First the evaluation of the filled in application dossier
2. Next the assessment of the international value of the geological sites of the applicant area by scientific professionals.
3. Followed by a field evaluation mission
4. And at last the approval by the UNESCO Global Geoparks council (UN, 2016).
The evaluation is done by experts and professionals which could make the application for becoming a Global Geopark more subjective (Bruschi and Cendrero, 2005; UNESCO, 2017b). The use of independent criteria of geodiversity and a GDI has not been mentioned in this process. It would improve the
application process of UNESCO to implement creating a GDI of the applicant park before the experts or professionals will go into the field. A proposed application process in summary would be after the addition of the GDI:
Application process:
1. Fill in application dossier of UNESCO 2. Assessment of international value
3. Create and analyze the Geodiversity Index of the applicant Geopark 4. Field evaluation by expert’s and verification of the GDI maps 5. Approval from the council
The creation and analyzation of the Geodiversity Index will reveal clusters of geodiversity which could be interesting spots to conserve. Moreover, management of the park can be based on the patterns of diversity found in the GDI. Therefore, it can give scientist study sites and points for education. This gives
the experts an opportunity to objectively evaluate the applicant Geopark before they visit the field. To eventually go into the field more prepared.
4.3 Limitations and Further research
There could have been experimented with more or different sub-indices. The input datasets where chosen because it was digitally available and appropriate for calculating a GDI of Fiordland National Park. The GDI reflects the quality of the input data, so differences in the classification system of an input map can have consequences for the GDI. For example, a geomorphology map can classify a mountainous area or can classify all individual mountains.
If all Geoparks would be covered by a GDI map, the connections between Geoparks in the Global
Geoparks Network (GGN) could be tightened (Benito-Calvo et al., 2009; Pellitero et al., 2011). Therefore, the development of a GDI mapping system for Geoparks could potentially be used to correlate
geodiversity and biodiversity indices (Parks and Mulligan, 2010). As geodiversity is an important driver for biodiversity, this can provide an efficient tool to conserve biodiversity as well as geodiversity in the same area of focus (Anderson and Ferree, 2010; Hjort and Luoto, 2010). As a suggestion, further research could connect the GDI of the Fiordland to a biodiversity index of this National park and investigate the indigenous plants and biotic factors with regard to the Gondwanaland resemblance.
5. Conclusion
based on the results of this study, the following conclusions can be made, that answer the research question. The most appropriate cell size for calculation of the GDI in the light of the input data is 750 meter. The most suitable equation of the GDI that should be applied for the Fiordland National Park is: GDI = Gdi + Gmdi + Sldi+ Hdi + Edi + Sdi * (0.5). The resulting geodiversity map should be interpreted in terms of clusters of High and Very high geodiversity patterns, and compared to the original patterns of the input data, to fully understand its meaning. These clusters could be interesting spots to conserve and management of the park can be based on the patterns of diversity. It is recommended that a GDI should be developed for a potential Geopark to enable objective evaluation of geo-factors by experts of UNESCO, before they visit the field. The aim is achieved by creating a GDI for the Fiordland National Park and based on this experience and literature review recommendations are given to UNESCO. This all to help enhance the preservation of geodiversity in natural protected areas in the future.
References
Benito-Calvo, A., Perez-Gonzalez, A., Magri, O. and Meza, P. (2009) Assessing regional geodiversity: the Iberian Peninsula. Earth Surface Processes and Landforms. 34, 1433-1445.
Bray, J. (2016). Assessing geodiversity in the Netherlands using GIS (scale issues) (Unpublished BSc thesis). University of Amsterdam, Amsterdam, The Netherlands.
Bruschi, V., M., Cendrero A. (2005). Geosite evaluation; can we measure intangible values? Il Quaternario, 18/1, 291-304.
Cody, A.D. (2007). Geodiversity of geothermal fields in the Taupo Volcanic Zone. DOC Research and Development Series 281. Department of Conservation, Wellington. 70p.
Crozier, M., J. (2010). Deciphering the effect of climate change on landslide activity: A review. Geomorphology. 124(3/4), 260-267.
Davey, F., J., Smith, E., G., C. (1983). The tectonic setting of the Fiordland region south-west New Zealand. Geophys. J. R. astr. SOC. 72, 23-38.
Department of Conservation (n.d.). Fiordland National Park. Retrieved on 22-06-2017 from: http://www.doc.govt.nz/parks-and-recreation/places-to-go/fiordland/places/fiordland-national-park/ Department of Conservation (2017). Milford Road/Milford Sound area. Retrieved on 01-07-2017 from: http://www.doc.govt.nz/milfordroad
Garnock‐Jones, P., J., Bayly, M., J., Lee, W., G., Rance, B., D. (2000). Hebe arganthera (Scrophulariaceae), a new species from calcareous outcrops in Fiordland, New Zealand, New Zealand Journal of Botany, 38:3, 379-388, DOI: 10.1080/0028825X.2000.9512690
Giardino, M., Bacenetti, M., Perotti, L., Giordano, E., Ghiraldi, L., Palomba, M. (2012). Geodiversity and geohazards of the Susa Valley (W-Alps, Italy): combining scientific research and new technologies for enhanced knowledge and proactive management of geoheritage in mountain regions. EGU General Assembly 2013, held 7-12 April, 2013 in Vienna, Austria, id. EGU2013-7204.
Gray, M. (2004). Geodiversity. Valuing and conserving abiotic nature. Chichester: John Wiley and Son. Gray, M., Gordon, J.E. and Brown, E.J. (2013). Geodiversity and the Ecosystem Approach: The Contribution of Geoscience in Delivering Integrated Environmental Management. Proceedings of the Geologists’ Association, Geoconservation for Science and Society. 124, 659-730.
Hancox, G., T., Perrin., N., D. (2009). Green Lake Landslide and other giant and very large postglacial landslides in Fiordland, New Zealand. Quaternary Science Reviews, 28, 1020-1036.
Hengl, T. (2006) Finding the right pixel size. Computers and Geosciences 32(9), 1283-1298.
Hewitt, A., E. (1998). New Zealand soil classification. 2nd ed. Landcare Research Science Series 1. Lincoln, Manaaki Whenua Press.
Hjort, J. and Luoto, M. (2010). Geodiversity of high-latitude landscapes in northern Finland. Geomorphology. 115, 109-116.
Kozlowski, S. (1999). Programme of geodiversity conservation in Poland. Polish Geological Institute Special Papers. 2, 15- 18.
Kozłowski, S. (2004) Geodiversity. The concept and scope of geodiversity. Przeglad Geologiczny 52(8/2), 833-883. Parks, K.E. and Mulligan, M. (2010). On the relationship between a resource based measure of geodiversity and broad scale biodiversity patterns. Biodiversity and Conservation. 19(9), 2751-2766.
Reynard, E., Coratza, P., 2007. Geomorphosites and geodiversity: a new domain of research. Geographica Helvetica. 62, 138–139.
Panizza, M., Piacente, S. (2008). Geomorpghosites and Geotourism. Revisita Geográfica Acadêmica, 2 (1), 5-9. Pellitero, R., Gonzalez-Amuchastegui, M.J., Ruiz-Flaño, P. and Serrano, E. (2011). Geodiversity and geomorphosites assessment applied to a natural protected area: the Ebro and Rudron gorges Natural Park (Spain). Geoheritage. 3(3), 163-174.
Prosser, C. ,D., Burek, C.,V., Evans, D.,H., Gordon, J.,E., Kirkbride, V.,B., Rennie, A.,F., Walmsley, C.,A. (2010). Conserving Geodiversity Sites in a Changing Climate: Management Challenges and Responses. Geoheritage. 2, 123– 136.
Räsänen, A., Kuitunen, M., Hjort, J., Vaso, A., Kuitunen, T., Lensu, A. (2016). The role of landscape, topog- raphy, and geodiversity in explaining vascular plant species richness in a fragmented landscape. Boreal Env. Res. 21: 53–70.
Reynard, E., Fontana, G., Kozlik, L., Scapozza, C. (2007) A method for assessing the scientific and additional values of geomorphosites. Univerisy of Lausanne.
Ruban, D., A. (2005). Quantification of geodiversity and its loss. Proceedings of the Geologists’ Association. 121, 326-333.
Seijmonsbergen A., Sevink, J., Cammeraat, L., Recharte, J. (2010). A potential geoconservation map of the Las Lagunas area, northern Peru. Environmental Conservation, 37: 107-115.
Seijmonsbergen A. C., De Jong, M. G. G., Hagendoorn, B., Oostermeijer, J. G. B., Rijsdijk, K. F. (in press).
Geodiversity mapping in alpine areas. Institute for Biodiversity and Ecosystem Dynamics, University of Amsterdam, Amsterdam, The Netherlands [unpublished].
Serrano, E., Ruiz-Flaño, P. (2007). Geodiversity: a theoretical and applied concept. Geographica Helvetica. 62(3), 140-147.
Tabachnick, B. G., and Fidell, L. S. (1996). Using multivariate statistics (3rd ed.). New York: HarperCollins.
Turnbull, I.M., Allibone, A.H., Jongens, R. (2010). Geology of the Fiordland area: scale 1:250,000. Lower Hutt: GNS Science. Institute of Geological and Nuclear Sciences. 1:250,000 geological map 17. 97 p. + 1 folded map
Melelli, L. (2014). Geodiversity: a new quantitative index for natural protected areas enhancement. GeoJournal of Tourism and Geosites. 13(1), 2-12.
United Nations (2015). Global Geoparks Network (GGN) Guidelines for Global Geopark Field Inspections and the Preparation of Evaluation / Revalidation Reports. United Nations Educational, Scientific and cultural Organization. Retrieved from: http://globalgeoparksnetwork.org/wp-content/uploads/2015/03/EVALUATORS-Conduct-Guidelines-2015.pdf
United Nations (2016). UNESCO Global Geoparks. United Nations Educational, Scientific and cultural Organization. Retrieved from http://unesdoc.unesco.org/images/0024/002436/243650e.pdf
UNESCO. (2017a). Te Wahipounamu – South West New Zealand. Retrieved on 6-09-2017 from http://whc.unesco.org/en/list/551/
UNESCO. (2017b). Statutes and Operational Guidelines for UNESCO Global Geoparks. Retrieved on 15-06-2017 from: http://www.unesco.org/new/fileadmin/MULTIMEDIA/HQ/SC/pdf/IGGP_UGG_Statutes_Guidelines_EN.pdf
UNESCO. (2017c). UNESCO Global Geoparks. Retrieved on 15-06-2017 from: http://www.unesco.org/new/en/natural-sciences/environment/earth-sciences/unesco-global-geoparks/
University of Otago (n.d.). Tectonic setting of New Zealand: astride a plate boundary which includes the Alpine Fault. Retrieved on 03-6-2017 from:
http://www.otago.ac.nz/geology/research/structural-geology/alpine-fault/nz-tectonics.html
Vregelaar, ten M. (2015). Developing a geodiversity index for the Netherlands (Unpublished BSC thesis). University of Amsterdam, Amsterdam, The Netherlands.
Walcott, R. I. (1998). Modes of oblique compression: Late Cenozoic tectonics of the south island of New Zealand. Rev. Geophys. 36(1), 1–26, doi: 10.1029/97RG03084.
Wing, S. R., Jack, L. (2014). Fiordland: the ecological basis for ecosystem management. New Zealand Journal of Marine and Freshwater Research. 48:4, 577-593, DOI: 10.1080/00288330.2014.897636
Zwoliñski, Z., Stachowiak, J. (2012). Geodiversity map of the Tatra National park for geotourism. Quastiones Geographicae. 31(1), 99-106.
The soil classification related to the abbreviations in the Soil map legend can be seen in a document called: LRIS
Data Dictionary - v3, which can be retrieved from:
https://lris.scinfo.org.nz/layer/79-fsl-new-zealand-soil-classification/
Appendix D:
Correlation matrices
CORRELATION
MATRIX – Equation 1 GDI 500 m Geology Geomorphology Soil Hydrology Elevation Slope GDI 500 m 1 0,37539 0,65134 0,54432 0,30267 0,52213 0,63927 Geology 0,37539 1 0,13427 0,61921 0,11645 -0,07500 0,04306 Geomorphology 0,65134 0,13427 1 0,33183 0,07266 0,29215 0,32542 Soil 0,54432 0,61921 0,33183 1 0,09120 0,07571 0,17211 Hydrology 0,30267 0,11645 0,07266 0,09120 1 -0,08802 0,09921 Elevation 0,52213 -0,07500 0,29215 0,07571 -0,08802 1 0,46186 Slope 0,63927 0,04306 0,32542 0,17211 0,09921 0,46186 1 CORRELATION
MATRIX – Equation 1 GDI 750 m Geology Geomorphology Soil Hydrology Elevation Slope GDI 750 m 1 0,29920 0,73576 0,61543 0,40881 0,51771 0,94459 Geology 0,29920 1 0,10426 0,42923 0,08504 -0,06138 0,29089 Geomorphology 0,73576 0,10426 1 0,36383 0,12543 0,36049 0,72753 Soil 0,61543 0,42923 0,36383 1 0,12887 0,12875 0,60422 Hydrology 0,40881 0,08504 0,12543 0,12887 1 0,01730 0,40094 Elevation 0,51771 -0,06138 0,36049 0,12875 0,01730 1 0,57109 Slope 0,94459 0,29089 0,72753 0,60422 0,40094 0,57109 1 CORRELATION
MATRIX – Equation 1 GDI 1000m Geology Geomorphology Soil Hydrology Elevation Slope GDI 1000 m 1 0,42632 0,68663 0,59411 0,41910 0,59089 0,68511 Geology 0,42632 1 0,12614 0,65884 0,08192 0,00844 0,17605 Geomorphology 0,68663 0,12614 1 0,32373 0,15831 0,38587 0,35085 Soil 0,59411 0,65884 0,32373 1 0,13896 0,17817 0,24432 Hydrology 0,41910 0,08192 0,15831 0,13896 1 0,09507 0,23269 Elevation 0,59089 0,00844 0,38587 0,17817 0,09507 1 0,39679 Slope 0,68511 0,17605 0,35085 0,24432 0,23269 0,39679 1
CORRELATION
MATRIX – Equation 2 GDI 500 m Geology Geomorphology Soil Hydrology Elevation Slope GDI 500 m 1 0,47419 0,70140 0,62995 0,36714 0,46154 0,56948 Geology 0,47419 1 0,13427 0,61921 0,11645 -0,07500 0,15113 Geomorphology 0,70140 0,13427 1 0,33183 0,07266 0,29215 0,29030 Soil 0,62995 0,61921 0,33183 1 0,09120 0,07571 0,21358 Hydrology 0,36714 0,11645 0,07266 0,09120 1 -0,08802 0,14855 Elevation 0,46154 -0,07500 0,29215 0,07571 -0,08802 1 0,32824 Slope 0,56948 0,15113 0,29030 0,21358 0,14855 0,32824 1 CORRELATION
MATRIX – Equation 2 GDI 750 m Geology Geomorphology Soil Hydrology Elevation Slope GDI 750 m 1 0.48676 0.75156 0.65932 0.4127 0.49929 0.93075 Geology 0.48676 1 0.14642 0.62853 0.10921 -0.04249 0.41946 Geomorphology 0.75156 0.14642 1 0.36383 0.12543 0.36049 0.72753 Soil 0.65932 0.62853 0.36383 1 0.12887 0.12875 0.60422 Hydrology 0.4127 0.10921 0.12543 0.12887 1 0.0173 0.40094 Elevation 0.49929 -0.04249 0.36049 0.12875 0.0173 1 0.57109 Slope 0.93075 0.41946 0.72753 0.60422 0.40094 0.57109 1 CORRELATION
MATRIX – Equation 2 GDI 1000m Geology Geomorphology Soil Hydrology Elevation Slope GDI 750m 1 0.4968 0.73063 0.65338 0.44946 0.56075 0.53843 Geology 0.4968 1 0.16839 0.65906 0.09488 0.01301 0.03341 Geomorphology 0.73063 0.16839 1 0.37467 0.189 0.42877 0.43767 Soil 0.65338 0.65906 0.37467 1 0.15445 0.1988 0.21884 Hydrology 0.44946 0.09488 0.189 0.15445 1 0.13524 0.24401 Elevation 0.56075 0.01301 0.42877 0.1988 0.13524 1 0.70859 Slope 0.53843 0.03341 0.43767 0.21884 0.24401 0.70859 1
CORRELATION
MATRIX – Equation 3 GDI 500 m Geology Geomorphology Soil Hydrology Elevation GDI 500 m 1 0,50893 0,72072 0,65950 0,39525 0,43489 Geology 0,50893 1 0,13427 0,61921 0,11645 -0,07500 Geomorphology 0,72072 0,13427 1 0,33183 0,07266 0,29215 Soil 0,65950 0,61921 0,33183 1 0,09120 0,07571 Hydrology 0,39525 0,11645 0,07266 0,09120 1 -0,08802 Elevation 0,43489 -0,07500 0,29215 0,07571 -0,08802 1 CORRELATION
MATRIX – Equation 3 GDI 750m Geology Geomorphology Soil Hydrology Elevation GDI 750 m 1 0.50911 0.74646 0.67614 0.41809 0.49089 Geology 0.50911 1 0.14642 0.62853 0.10921 -0.04249 Geomorphology 0.74646 0.14642 1 0.36383 0.12543 0.36049 Soil 0.67614 0.62853 0.36383 1 0.12887 0.12875 Hydrology 0.41809 0.10921 0.12543 0.12887 1 0.0173 Elevation 0.49089 -0.04249 0.36049 0.12875 0.0173 1 CORRELATION
MATRIX – Equation 3 GDI 1000m Geology Geomorphology Soil Hydrology Elevation GDI 1000 m 1 0.50991 0.74354 0.67702 0.44579 0.52933 Geology 0.50991 1 0.16839 0.65906 0.09488 0.01301 Geomorphology 0.74354 0.16839 1 0.37467 0.189 0.42877 Soil 0.67702 0.65906 0.37467 1 0.15445 0.1988 Hydrology 0.44579 0.09488 0.189 0.15445 1 0.13524 Elevation 0.52933 0.01301 0.42877 0.1988 0.13524 1
CORRELATION MATRIX – Equation 4
GDI 500m Geology Geomorphology Soil Hydrology
GDI 500 m 1 0.583 0.69787 0.70772 0.47118 Geology 0.583 1 0.13427 0.61921 0.11645 Geomorphology 0.69787 0.13427 1 0.33183 0.07266 Soil 0.70772 0.61921 0.33183 1 0.0912 Hydrology 0.47118 0.11645 0.07266 0.0912 1 CORRELATION MATRIX – Equation 4
GDI 750m Geology Geomorphology Soil Hydrology
GDI 750m 1 0,59446 0,71825 0,72730 0,48543 Geology 0,59446 1 0,14642 0,62853 0,10921 Geomorphology 0,71825 0,14642 1 0,36383 0,12543 Soil 0,72730 0,62853 0,36383 1 0,12887 Hydrology 0,48543 0,10921 0,12543 0,12887 1 CORRELATION MATRIX – Equation 4
GDI 1000m Geology Geomorphology Soil Hydrology
GDI 1000 m 1 Geology 0.50991 1 Geomorphology 0.72891 0.12614 1 Soil 0.67709 0.65884 0.32373 1 Hydrology 0.45317 0.08192 0.15831 0.13896 1
