Amphibians in the Region of Murcia (SE Iberian peninsula ... - Raco

36 Egea–Serrano et al.Luiselli, 2000). According to these authors, thisstatistical approach allows studied species to begrouped in a more suitable way than univariatetechniques since relations between variables canbe established. Anuran and urodele species wereanalysed together because no information is availableconcerning the main differences between thesetwo groups in relation to their biology and sensitivityto habitat degradation. The multifactorial analysisused was a multiple correspondence analysis(MCA). This statistical technique allows informationprovided by original data to be reduced to twodimensions which explain most data variance, andassigns a new coordinate to each case for eachdimension extracted by the analysis (Visauta, 1998).According to the values obtained for each dimension,species have been assigned to one of thefollowing categories, ranging from low to high riskof extinction: 1. Species showing positive values forboth dimensions (low risk of extinction); 2. Speciesshowing positive values for dimension 1 and negativevalues for dimension 2 (low–medium risk ofextinction); 3. Species showing negative values fordimension 1 and positive values for dimension 2(medium–high risk of extinction); 4. Species showingnegative values for both dimensions (high riskof extinction).To establish priority conservation areas from thepoint of view of the amphibian species present, thesurface area of the Region of Murcia was dividedinto a 5 x 5 km UTM grid. The number of amphibianspecies present for each square was determinedaccording to information presented by Egea–Serranoet al. (2005b, 2005c). Additionally, the proportion ofsquares occupied by each species was calculatedin relation to the number of 5 x 5 UTM squares intowhich the study area was divided. This procedureallowed to estimate species extension in the studyarea, establishing an index of area occupation inMurcia (D) with five categories ranging from high tolow presence: 1. Species present in > 30% of thesurface of the region. This area corresponds to thearea of occupancy of a species considered as NearThreatened or Least Concern according to UICNcategories (UICN, 2001). 2. Species present in 10–30% of the surface of the region. This area correspondsto the area of occupancy of a specieswhose risk of extinction can be considered intermediatebetween Near Threatened or Lleast Concernand Vulnerable categories, according to UICN criteria(UICN, 2001). 3. Species present in 5–10% ofthe surface of the region. This area corresponds tothe area of occupancy of a species consideredVulnerable according to UICN categories (UICN,2001). 4. Species present in 1–5% of the surface ofthe region. This area corresponds to the area ofoccupancy of a species whose risk of extinction canbe considered intermediate between Vulnerable andEndangered categories, according to UICN criteria(UICN, 2001). 5. Species present in < 1% of thesurface of the region. This area corresponds to thearea of occupancy of a species considered Endangeredaccording to UICN categories (UICN, 2001).Considering the calculated amphibian distributiondata, number of species per square and thepreviously calculated risk of extinction for eachspecies, a biological value was calculated for each5 x 5 km square through the expression:(MCA i+D i)+Spp jwhere MCA iis the risk of extinction for species i, D iis the distribution of species i in the Region ofMurcia, and Spp jthe number of amphibian speciesfor j square.Squares showing values higher than the 75 thpercentile for this index were selected as priorityconservation squares.Statistical analysis were performed with theSPSS ® statistical package.ResultsThe scores for the independent variables for theamphibian species in the Region of Murcia arepresented in table 2. The results provided by multiplecorrespondence analysis have enabled identificationof three groups of species (fig. 1). Table 3shows the scores for each variable in each dimensionextracted by the MCA. Breeding habitat combinedthe highest value for dimension 1 (0.770) andthe lowest for dimension 2 (0.067), whereas speciespresence presented the lowest value for dimension1 (0.682) and the highest for dimension 2(0.911). This implies that the main variables arrangingspecies through dimension 1 and dimension2 in the Region of Murcia are, respectively,breeding habitat and species presence.Table 4 shows the values obtained for variablesrisk of extinction and extension in the study area foreach amphibian species.The 5 x 5 km UTM squares showing biologicalvalue indices higher than the 75 th percentile wereconsidered as priority conservation areas. The totalnumber of such areas added up to 103 (fig. 2) andrepresented 16% of the surface of the Region ofMurcia.DiscussionAlthough Andreone & Luiselli (2000) indicate thatboth univariate and multivariate methodology sufficientlycharacterise the conservation status of aspecies group, multivariate analysis alone wasused in the present study as univariate analysis isconsidered to have the disadvantage of not establishingrelationships between variables and thereforenot realistically ranking the studied speciesaccording to their risk of extinction.The results obtained show that the speciesexposed to higher risk of extinction are thosewhich depend on the presence of permanent waterbodies to complete their larval development andwhich, in addition, show a restricted distribution