A PROPOSAL OF QUANTITATIVE GEODIVERSITY EVALUATION MODEL ON THE EXAMPLE OF UPPER PIVKA KARST , SLOVENIA

The article presents a new method for geodiversity evaluation, which was evaluated in the Upper Pivka karst area. The method is based on entire area evaluation on the basis of spatial distribution of geodiversity elements and terrain ruggedness. The method was developed for the purpose of objective and systematic geodiversity evaluation on various sites. In this article, we applied the method in the Upper Pivka area, where it demonstrated to be very useful.


INTRODUCTION
The diversity of nature is composed of the biotic and abiotic components.The first occurrences of natural diversity evaluation were based on geomorphological and geological elements, which also reflected in the establishing of the first geological reserve (Siebengebirge in the former Prussia in 1836) and the first national park in the world (Yellowstone in 1872) (Gray, 2013).Current nature diversity studies are paying more attention to biotic component of nature, as a result of excessive exploitation of natural resources and consequent extinction of plant and animal species, as well as habitats (Pettersson, Keskitalo, 2013).After the signing of the Convention on Biological Diversity in Rio de Janeiro in 1992, the number of studies in this topic rapidly increased; there have been numerous studies for evaluation, protection, and conservation of biodiversity.The latter refers to genetic (conserving genetic diversity), species (reducing species loss), and ecosystem diversity (maintaining and protecting habitats), as well as protecting their interconnections.
An excessive biocentric approach to nature protection almost entirely overlooked the abiotic aspect of nature.Interest in protection and evaluation of abiotic nature has increased in the last two decades (Gray, 2013;Melelli, 2014).The modern term geodiversity encompasses the diversity of geologic (rocks, minerals, fossils) and geomorphologic elements (forms and processes) and soils.The revival of interest for evaluation of abiotic nature is not only indicated by expert articles and the search for the most suitable methods of abiotic nature evaluation.There is also increasing interest for so-called geotourism, which is based on the combination of geologic and geomorphologic interpretation of landscape features and recreation (Necheş, 2016), with an increasing number of geoparks, which are a direct result of increased interest in abiotic nature elements.
With purpose for geodiversity evaluation, various methods had been developed, which evaluate geodiversity in different manner.All methods use a two-stage evaluation of scientific and additional values.Scientific criteria are generally based on the scientific understanding of a relief form or process, while additional values define the utilization and managerial potential (Pereira, Pereira, Caetano Alves, 2007;Reynard, Coratza, 2007;Zouros, 2007;Reynard, 2009;Gray, 2013).The evaluation methods differ significantly also in the way of acquiring information on geodiversity elements.Simpler methods evaluate individual points or areas, while modern methods consider the entire area under evaluation, and define the areas with a higher geodiversity index (Melelli, 2014;Stepišnik, Repe, 2015).
The purpose of this article is to present a new, semi-automated quantitative method of geodiversity evaluation.The evaluation procedure is predominantly based on the analysis of digital spatial data in geographic information systems.Most attempts at evaluation of abiotic nature diversity so far have been based on an evaluation of individual points and/or connected areas of geodiversity elements.Our proposed model assesses the entire study area.
The fundamental problem of geodiversity evaluation is the subjective element of assessment, which is generally the result of loosely defined evaluation criteria and the complexity in choosing individual geodiversity elements.Final values are generally the result of the assessor's own judgment, i.e. clearly subjective (Melelli, 2014;Stepišnik, Repe, 2015).The goal of our method is to eliminate, as much as possible, the subjective element of assessment from the evaluation.The only part of the procedure that includes subjective elements is the selection of geodiversity elements.For this reason, the identification of geodiversity elements is as simplified as possible and limited to basic elements, which can be identified by any assessor with a basic education in geography and geology.
In addition to the geodiversity elements, the method considers surface fragmentation, which is also an important factor of abiotic nature diversity (Hjort, Luoto, 2010;Ruban, 2010;Gray, 2013;Pereira, Pereira, Caetano Alves, 2013;de Paula Silva, Rodrigues, Pereira, 2014;Melelli, 2014;Stepišnik, Repe, 2015;Necheş, 2016).The final calculation of geodiversity index combines information on locations of different geodiversity elements and the terrain ruggedness.The method of determining the geodiversity index is based on the Block statistics tool in the ArcGis software package.
The method has already been successfully used on the paleovolcanic area of Kratovo in Macedonia (Trenchovska, 2016).In the study, we applied the method in the southern part of the Pivka Basin, or Upper Pivka, which is one of the most diverse areas in Slovenia in terms of geomorphology and hydrology.We hope to present the objectivity and usefulness of this model for karst-type terrain regions.

MATERIALS AND METHODS
To inventory and evaluate the geodiversity, we used a digital LiDAR elevation model with a 1x1 m resolution (ARSO, 2015) and a digital morphographic map of geodiversity elements.For spatial analyses and the geodiversity index calculations, we used Esri's Arc Map software, version 10.3.1.All analyses and index calculations were performed for a spatial unit in the shape of a 200 x 200 m square.The size of the spatial unit was adapted to the size of the entire studied area of Upper Pivka.The chosen method of allocation allows objectivity in taking inventory, in the analysis of created data, and in the comparison of final values.
The inventory and geodiversity evaluation model is composed of three main phases.In the first phase, geodiversity elements are identified and spatially documented on the basis of terrain mapping, scientific literature (Serrano, Ruiz-Flaño, 2007;Gray, 2013), cartographic material of various scales (1:5.000;1:25.000), and digital orthophotos (GURS, 2014).Our model takes into account the most important geomorphologic and hydrologic geodiversity elements.For the entire study area, we used the National topographic map, 1:25.000scale to map three different types of geomorphologic landforms (larger karst hollows, pocket valleys, conical hills) and three different types of hydrologic landforms (springs, rivers, intermittent lakes).The types were later digitized and vector layers of specific types were prepared.Of the topographic elements, we selected terrain ruggedness, which was calculated as a terrain ruggedness index on the basis of an analysis of a digital LiDAR elevation model (Riley, DeGloria, Elliot, 1999).The results of the first phase were vector layers of geodiversity elements and a terrain ruggedness index, which represent the basis for further analysis.
The second phase of the model includes the inventory of geodiversity elements for each spatial unit.First, we transformed vector layers of geodiversity elements into raster layers with a cell size of 1 x 1 m; this cell size matches with the cell size of the digital elevation model.The inventory was performed on the basis of a statistical analysis of the maximum, using the Block statistics tool (Neighborhood toolset -maximum value).We determined the presence or absence of an element within a spatial unit for each element separately.Next, we reclassified and performed a sum of the raster layers using the Raster calculator tool.We thus obtained the precise number of different elements occurring within the 1828 spatial units of equal size in the entire studied area.
The third phase represents the geodiversity index calculation using the equation (Serrano, Ruiz-Flaño, 2007), which we modified and adapted for the purpose of our model.
The modified equation is as follows (Trenchovska, 2016): Where Gd is the geodiversity index, Eg is the number of different geodiversity elements within spatial unit, and R is the terrain ruggedness index of a spatial unit.
The layers of terrain ruggedness index and the number of different geodiversity elements within spatial unit were multiplied using the Raster calculator tool to obtain the geodiversity index which ranges from 0 to 1220.07.The geodiversity index was classified into three natural Jenks' classes (Jenks, 1967).We thus obtained a low, medium, and high geodiversity index.For easier use and interpretation of results, we transformed classes into homogenous rounded units.Areas with a high index are geodiversity hotspots (Ruban, 2010).

TERRAIN CHARACTERISTICS OF UPPER PIVKA
The entire area of Pivka Basin is a lower terrain, predominantly composed of flysch rocks and partially carbonate rocks, over which flow the two main rivers, Pivka and Nanoščica, with their tributaries.The basin is enclosed on all sides by a higher karst terrain of high Dinaric plateaus Nanos, Hrušica, Javornik, and Snežnik in the north and east, and borders in the west on Slavenski ravnik, which transitions in the south into Taborski greben.The waters from the Pivka Basin flow into different parts of the surrounding areas, where they flow underground in different directions.The area represents the watershed between the Adriatic and Black Sea catchments, and due to the complex geological compositions the water is draining to both catchments.From geomorphological, geological, and hydrological perspective, it is one of the most complex units of Slovenian Dinaric karst.
Due to differences of surface formations and geological features, the Pivka Basin is divided into Upper Pivka (Zgornja Pivka), or Podsnežniška Pivka, and Lower Pivka (Spodnja Pivka), or Podnanoška Pivka.The Lower Pivka lies in the lower section of the Pivka drainage basin, north of Prestranek, and roughly includes the entire drainage basin of Nanoščica.It is composed primarily of flysch rocks; therefore, the fluvial relief developed in this area.The Upper Pivka includes the southern part of the Pivka Basin between Prestranek and Šembije.It is composed primarily of carbonate rocks, while the bottom of the basin and surrounding hollows are also covered by fluvial deposits (Slovenija: pokrajine in ljudje, 2001).
The area of Upper Pivka is a roughly flattened landscape, divided into the valley floor along the Pivka River and its tributaries, and the higher karst plain, which is dissected by various karst hollows and conical hills.It spans from Šembije in the south, borders on Taborski greben in the west and Javorniki in the east.Only in the north it is not bounded by a higher terrain, but opens up by Prestranek into the remaining part of the Pivka Basin, or Lower Pivka.The entire area is about 15 km long and up to 5 km wide, gently sloping down from the far southeastern part near the village of Koritnice, with an altitude of about 640 m, to Prestranek, with an altitude of about 520 m.The terrain of the Upper Pivka Valley can be divided into two larger morphological units: the valley bottom covered by various fluvial sediments (Figure 2A) and the higher positioned karst plain (Figure 2B), which includes the southern part of Upper Pivka between the surrounding slopes of Javorniki, Taborski greben, and the edges of Snežnik Mountain.A smaller continuous area of the flattened floodplain is located between the settlements of Koritnice, Bač, and Knežak.This is an area of sand-gravel deposits of temporary streams, flowing from the area beneath Velika Milanja, where a fluviokarst geomorphic system developed due to predominantly dolomite rock composition.A whole line of prominent conical hills rises above this area.
The largest continuous floodplain in the Upper Pivka area lies along the Pivka River itself, from its spring in the village of Zagorje to the northern edge of the area near Prestranek.The wide floodplain, covered by fine-grained sediments of Pivka River, transitions into narrow and shallow canyons in some locations.The area also includes numerous karst springs of Pivka tributaries; some of them are located in characteristic pocket valleys.
The higher, bedrock terrain, which comprises the largest portion of Upper Pivka, is completely karstified.It is covered by numerous sinkholes, conical hills, and other smaller karst landforms.It is also characterized by numerous larger and deeper hollows.Bottoms of some hollows reach the level of the epiphreatic zone and therefore occasionally are inundated.

EVALUATING GEODIVERSITY OF UPPER PIVKA
Geodiversity elements were mapped in detail in the field, also by using remote sensing.Based on mapping, we determined six different geodiversity elements, of which three types were hydrological and three types were geomorphological.These elements are the most important for defining the geodiversity index, and they also are so basic that can be documented in the same manner by multiple assessors.Thus, we have eliminated the subjective aspect of assessing to the greatest degree possible.
In the study area, we identified a total of 89 geodiversity elements, which are discontinuously distributed (Figure 3).The most characteristic geomorphological landforms in the study area are larger karst hollows (36) and conical hills (28).Two smaller pocket valleys are also present in the area.Some important springs (5) and the surface flow of Pivka River with its tributaries are located in the western part of the area.A special feature of the shallow karst of Upper Pivka are undoubtedly the intermittent lakes (17).The terrain ruggedness index was calculated on the basis of statistical analyses (Block statistics) of the digital LiDAR elevation model and using Riley's equation, which can be used for a quantitative display of topographic ruggedness of areas of different sizes (Riley, DeGloria, Elliot, 1999).The calculation was done for a spatial unit of a 200 x 200 m block, in order to match with the results of other statistical analyses of geodiversity elements and to generalize the precision of LiDAR data.Index values ranged from 34.12 to 406.69.Spatial units with high index values are located in the eastern part of the area, where karst plain gradually rises towards the slopes of Javorniki, in the far southern part of the area on the outskirts of Snežnik Mountain and on the slopes of larger karst hollows.Low ruggedness index values were predominantly in the western part of the studied area, on the levelled terrain along the Pivka River and its tributaries, and on the extensive fluvial karst plain in the southeast (Figure 4).
The entire 75 km 2 of the study area was divided into 1828 spatial units (blocks) of equal size.Using a statistical analysis of maximum (Block statistics -maximum value) and summing (Raster calculator), we determined the exact number of different geomorphologic and hydrologic geodiversity elements for each block, which ranged between 0 and 3 (Figure 4).Most spatial units (1321) did not contain a single geodiversity element.Only one element was present in 354 units, two elements in 146 units, and three elements in 7 units.The results obtained with the new quantitative model indicate that over half (73.67%) of the study area falls within the low geodiversity index class.This class is predominant in areas where there the number of different geodiversity elements is zero and the area has a low terrain ruggedness index.These are valley bottoms with low elevation and low slope gradient.
The medium geodiversity index covers 18.98% of the study area.It occurs in area with one or two different geodiversity elements and where the terrain ruggedness index has a medium value.This index is mostly distributed along the contact of karst plain and the valley floor, and in areas of larger karst hollows.
Areas with high geodiversity index represent 7.35% of the Upper Pivka area.Such areas with a high geodiversity index can be defined as geodiversity hotspots (Ruban, 2010).These areas have three different geodiversity elements occurring within the basic spatial unit cells, and also have a high terrain ruggedness index.
In the study area of Upper Pivka, there are seven continuous areas with a high geodiversity index (Figure 5).The northernmost area is located in the area of Jeredovce.This is an elongated karst hollow with an intermittent lake.There are also some conical hills around the hollow, which together with the high terrain ruggedness index produces relatively high geodiversity index values.Towards the south, areas with a high geodiversity index because of similar geomorphological, hydrological, and terrain characteristics are also located around Petelinje, Palčje, Bač and Šembije Lakes; all these areas have large karst hollows with intermittent lakes surrounded by conical hills.A smaller continuous area with a high geodiversity index is located about 700 m southeast from the spring of Pivka, around Kalec Lake.The high geodiversity index of this area is defined by a high terrain ruggedness index and the karst hollow with the intermittent lake, as well as the vicinity of the surface flow of the Pivka River.
The largest continuous area with a high geodiversity index is the part of the karst plain to the east of the village Parje.This area includes seven larger karst hollows, six of which have intermittent lakes.There are also karst springs, a conical hill, and surface flow of the Pivka River with its tributaries close to this area.The high diversity of geomorphologic and hydrologic elements, together with a relatively high terrain ruggedness index, gives to this area a high geodiversity index.

DISCUSSION AND CONCLUSION
The concept of geodiversity was introduced in the last two decades due to increasing interest in abiotic nature elements (Gray, 2013;de Paula Silva, Rodrigues, Pereira, 2014).Geodiversity evaluation is important for the managing of specific areas from the perspective of nature protection, geotourism, and education.Geodiversity is expressed as a value of an area that can be defined by qualitative and/or quantitative indicators.The greatest problem of qualitative methods is the subjectivity of assessors.When evaluating geodiversity, assessors apply their personal views towards abiotic nature.Study results therefore differ significantly and are highly subjective.Qualitative indicators are also neither suitable for comparative studies, nor for studies done by more than one assessor (Ruban, 2010).
A much greater objectivity of the results can be achieved using a quantitative approach (Hjort and Luoto, 2010;Ruban, 2010;Melelli, 2014).This approach allows comparison and combining data of various evaluation studies.Furthermore, a quantitative approach, which generally uses GIS tools, can effectively process a much larger quantity of data.Such an approach is extremely practical for evaluating larger areas or areas with a higher density of geodiversity elements.It also allows comparison of data for the same area during different time periods; we can thus evaluate a loss of geodiversity through time for a specific area.Using quantitative methods, we can also identify areas of high geodiversity index or geodiversity hotspots.
The method used in our study combines the spatial relationship of terrain ruggedness with geodiversity elements.The method is partially automated, but at the same time eliminates the subjective component of assessment using a very simple identification and documentation of geodiversity elements.The method was tested in the Upper Pivka area, where we determined seven different areas, which can be defined as geodiversity hotspots.The identified areas match the intermittent lakes, which are also currently

Figure 1 :
Figure 1: Location of the study area.

Figure 2 :
Figure 2: A. Karst plain with intermittent lakes in the hinterland of the Mišnik spring near Parje; the view towards the south (photo: U. Stepišnik).B. Floodplain by the spring of Pivka in Zagorje; the view towards the west (photo: U. Stepišnik).

Figure 4 :
Figure 4: Terrain ruggedness index and number of different geodiversity elements within a spatial unit.

Figure 6 :
Figure 6: Area of intermittent Palčje Lake is one of the high geodiversity index areas in Upper Pivka; view towards the north (photo: U. Stepišnik).

Table 1 :
Surface area and percentage of geodiversity index within Upper Pivka.