Naslov
Preliminary analyses results of forest plant diversity and distribution on Mt. Medvednica, Croatia

Autori
Jelaska, Sven D.

Vrsta, podvrsta i kategorija rada
Sažeci sa skupova, sažetak, znanstveni

Izvornik
GIS RESEARCH UK 2000 Conference Proceedings / Halls, Peter J. - York : University of York, 2000, 198-199

Skup
GIS Research UK 2000

Mjesto i datum
York, Ujedinjeno Kraljevstvo, 05.04.2000. - 07.04.2000

Vrsta sudjelovanja
Poster

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
flora; GIS; predictive model; regression

Sažetak
INTRODUCTION The mapping of Croatian flora should be one of the primary goal in the future Croatian botanical activities. The results achieved to date indicate that given the traditional way of mapping and the present number of active botanists this goal is far from being accomplished. The use of GIS in combination with statistical models to explain flora and vegetation distribution could provide a good basis for at least a simulated atlas of Croatian flora. In addition to this, large-scale principal goal, I wanted to demonstrate the applicability of GIS and modelling in the management of protected areas. Therefore, I have chosen Medvednica Nature Park as a test area for the analysis of flora distribution and predictive mapping. While some authors (Kadmon and Danin, 1997 ; Austin, 1998) use mean annual precipitation and temperature to explain and model the plant distribution on large-scale, in this research I concentrated on other abiotic factors similar to the approach of other small-scale surveys (McCune and Allen, 1985 ; Kazda, 1995). Because of the vicinity of the Nature Park to the city of Zagreb, the biggest industrial centre in the country, I focused on concentrations of base metals and some other chemical elements in the soil to interpret the distribution of the flora in the Nature Park. MATERIAL AND METHODS Along the central transverse profile of the Nature Park, determined by nine MTB 1/64 grid units, I defined 28 rectangular 50x50 meter plots of forest vegetation, on an altitudinal gradient from 195 to 954 meters above sea level. Using the differentially corrected GPS data I geo-referenced the plots. After establishing GIS for the surveyed transect I gathered information from already existing maps in paper (vegetation, geology, pedology) and digital (topography) format for every plot. During the field work I collected information about floristic compositions of the plots and concentrations of 35 chemical elements and acidity of the upper 10 centimetres of topsoil. Using the stepwise multiple linear regression the total plant species number dependence on the recorded abiotic factors was determined. For analysis of floristic composition of the plots I employed the Canonical Correspondence Analysis (Ter Braak, 1988). This way I extracted variables that best explained the variability in floristic composition of the surveyed area. With these variables I modelled the spatial distribution of single species. For all variables significant either for explaining plant species number and species composition I have made the interpolation on transect area that enable me to implement developed models on wider area predicting the spatial distribution of vascular plant diversity and particular species. RESULTS AND CONCLUSIONS The total number of plant species per 50x50 meters area was found to be positively correlated with the topsoil acidity and concentrations of cobalt, chromium and yttrium, while the correlation with magnesium concentration was negative. Variables that gave the best descriptions of the variability in floristic composition among the plots were altitude, soil acidity, concentrations of nickel, cobalt and potassium. Species that have their spatial variation sufficiently explained were mostly widely distributed species, which is concordant with findings of other authors (Cherrill et al., 1995). For rare species I applied the approach of overlapping layers of variables significant for explaining spatial distribution of species and found areas with environmental conditions similar to those on known localities (Sperduto and Congalton, 1996 ; Pfab and Witkowski, 1997). Here presented results from the survey of Medvednica Nature Park, with yet unpublished data, provides a framework for the management of protected and other areas. Further modification is needed, as well as enlargement of the sample. The incorporation of remote sensing data will increase the temporal and spatial accuracy of vegetation data. Hence sampling will be more objective. Additional large- and small-scale studies are needed in different climatic areas. Upon completing all the above-mentioned tasks it will be possible to work out a scheme for the production of an atlas of Croatian flora using GIS and remote sensing. REFERENCES Austin, M. P. (1998) An ecological perspective on biodiversity investigations: examples from Australian eucalypt forests. Ann. Missouri Bot. Gard. 85: 2-17 Cherrill, A. J., McClean, C., Watson, P., Tucker, K., Rushton, S. P. and Sanderson, R. (1995) Predicting the distributions of plant species at the regional scale: a hierarchical matrix model. Landsc. Ecol. 10(4): 197-207. Kadmon, R. and Danin, A. (1997) Floristic variation in Israel: a GIS analysis. Flora 192: 341-345 Kazda, M. (1995) Changes in alder fens following a decrease in the ground water table: results of a geographical information system application. J Appl Ecol 32: 100-110 McCune, B. and Allen, T. F. H. (1985) Will similar forests develop on similar sites? Can. J. Bot. 63: 367-376 Pfab, M. F. and Witkowski, E. T. F. (1997) Use of Geographical Information Systems in the search for additional populations, or sites suitable for re-establishment, of the endangered Northern province endemic Euphorbia clivicola. S. Afr. J. Bot. 63 (6): 351-355. Sperduto, M. B. and Congalton, R. G. (1996) Predicting rare orchid (Small Whorled Pogonia) habitat using GIS. Photogr. Eng. Remote Sensing 62(11): 1269-1279. Ter Braak, C. J. F. (1988) CANOCO - a FORTRAN program for canonical community ordination by [partial] [detrended] [canonical] correspondence analysis, principal components analysis and redundancy analysis (version 2.1). Technical report: LWA-88-02. Agricultural Mathematics Group, Wageningen.

Izvorni jezik
Engleski

Znanstvena područja
Biologija

POVEZANOST RADA