Effects of land-cover transformation and climate change - FCB-UJED

Loading...
Journal of Arid Environments 83 (2012) 1e9

Contents lists available at SciVerse ScienceDirect

Journal of Arid Environments journal homepage: www.elsevier.com/locate/jaridenv

Effects of land-cover transformation and climate change on the distribution of two endemic lizards, Crotaphytus antiquus and Sceloporus cyanostictus, of northern Mexico H. Gadsden a, *, C. Ballesteros-Barrera b, O. Hinojosa de la Garza a, G. Castañeda c, C. García-De la Peña c, J.A. Lemos-Espinal d a

Instituto de Ecología, A.C. Centro Regional Chihuahua, Miguel de Cervantes No. 120, Complejo Industrial Chihuahua, Chihuahua C.P. 31109, Mexico Universidad Autónoma Metropolitana-Iztapalapa, Av. San Rafael Atlixco No. 186, Colonia Vicentina C.P. 09340, Delegación Iztapalapa, D.F., Mexico Escuela Superior de Biología, Universidad Juárez del Estado de Durango, Av. Universidad s/n, Fraccionamiento Filadelfia, C.P. 35010 Gómez Palacio, Durango, Mexico d Laboratorio de Ecología, Unidad de Biotecnología y Prototipos, Facultad de Estudios Superiores Iztacala, UNAM, Apartado Postal 314, Avenida de los Barrios No. 1, Los Reyes Iztacala, Tlanepantla, Estado de México 54090, Mexico b c

a r t i c l e i n f o

a b s t r a c t

Article history: Received 19 July 2011 Received in revised form 20 March 2012 Accepted 22 March 2012 Available online xxx

Two species of lizards, Sceloporus cyanostictus and Crotaphytus antiquus, are restricted to small areas of rocky hills in the center of the Chihuahuan Desert, where land-cover transformation has increased dramatically in recent years and future climatic changes are expected to be severe. The current geographic distribution of each species was estimated by ecological niche modeling using the Maximum Entropy model (MaxEnt). A recent land-use map was used to determine areas where habitat has been transformed by human activities, and niche models were projected under one simulated climatic scenario and for two periods of time (2020 and 2050) to estimate their future potential distributions. Results indicate a high degree of anthropogenic habitat transformation within the distribution of C. antiquus, and a significant reduction of its current distribution is expected by 2050. For S. cyanostictus land-cover transformation is less severe, however a severe reduction of its current distribution is expected in the future because of climate changes. Despite the uncertainty involved, the general trends seem highly feasible and immediate conservation actions are recommended. Ó 2012 Elsevier Ltd. All rights reserved.

Keywords: Climatic change Coahuila Ecological niche modeling Lizard MaxEnt Reptiles

1. Introduction Natural fluctuations and local extinctions are frequent in reptilian populations; increasing data indicates an elevated decline worldwide during the last two decades (Pounds et al., 1999; Sinervo et al., 2010). Important factors appear to include habitat fragmentation, introduction of exotic species, pollution, diseases, parasitism, and global climate change (de Chazal and Rounsevell, 2009; Sinervo et al., 2010). Effects of global warming on declining of reptiles are relatively unexplored, but there is a suggestion that large impacts may occur on species that have temperaturedependent sex determination, such as crocodiles and some turtles (e.g., Ballesteros-Barrera et al., 2007; Godfrey et al., 1999).

* Corresponding author. Tel.: þ52 614 451 0905. E-mail addresses: [email protected] (H. Gadsden), [email protected] gmail.com (C. Ballesteros-Barrera), [email protected] (O. Hinojosa de la Garza), [email protected] (G. Castañeda), [email protected] gmail.com (C. García-De la Peña), [email protected] (J.A. Lemos-Espinal). 0140-1963/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved. doi:10.1016/j.jaridenv.2012.03.014

Mexico is considered one of the seven megadiverse countries in the world, holding about 10e15% of terrestrial species (Mittermeier and Goettsch, 1992), and is particularly significant in terms of reptilian diversity with the second highest number of species in the world (804), of which practically 50% are endemic and have limited distributions (Flores-Villela and Canseco-Márquez, 2004). This notable biodiversity is gravely threatened due to the accelerated rate of land-cover alteration in numerous regions of the country. Mexico has one of the highest deforestation rates in Latin America, an estimated 395 000 ha per year (FAO, 2005). Therefore, many reptilian species are considered to be at some level of risk by Mexican Official Norm (NOM-059-SEMARNAT-2010 see www.ine. gob.mx). Although an increase in biosphere reserves in Mexico has been designed to address this risk, numerous at-risk species continue unprotected: only 46% of the species considered to be endangered are protected by natural parks and reserves (SantosBarrera et al., 2004). A supplementary risk for numerous species is change in climatic regimes due to global warming (Pounds et al., 1999; Thomas et al.,

2

H. Gadsden et al. / Journal of Arid Environments 83 (2012) 1e9

2004). Since the last century, Earth’s climate has been changing at exceptional rates, with an increase in observed global temperature on the order of 0.6  C (IPCC, 1995). Predictions of future temperature increases are in the range of 2.5e5.4  C by 2100, which is much higher than the observed change in the 20th century (IPCC, 1995, 2001). It is predicted that severe climatic changes will affect the center of the Chihuahuan Desert in the coming decades, particularly in the period 2020e2050. According to the A2 scenario, an increase of around 2  C and a significant reduction of rainfall are expected in the region by 2050 (Ballesteros-Barrera et al., 2007). Species’ responses to the combined effect of land-cover transformation and climate change have been rarely studied despite being both key drivers of biodiversity change (de Chazal and Rounsevell, 2009). However, it is logical to suppose that species with higher environmental specialization and more restricted distributional ranges would be more vulnerable to changes in their habitat compared to wide-ranging generalist species. Recent distributional shifts have been recorded for diverse species of plants and animals, sometimes increasing and others reducing their area, but distribution limits normally have moved towards the poles or to higher altitudes (Sinervo et al., 2010), in severe cases causing the extinction of populations or species (Pounds et al., 1999). The Venerable Collard Lizard (Crotaphytus antiquus) and the Blue-spotted Spiny Lizard (Sceloporus cyanostictus) are two microendemic saxicolous species confined to the mountain islands of southwestern and southeastern Coahuila and western Nuevo León (Axtell and Webb, 1995; Price et al., 2010). According to Sinervo et al. (2010) island lizards of various species inhabiting the central plateau in northern Mexico are endangered in the short term due to global warming. It is therefore important to analyze ecological niche modeling (ENM) to detect current and future potential distribution of these species for the purpose of designing subsequent conservation plans best suited to climate change and anthropogenic degradation that are operating in their habitats. On the other hand the high endemism of lizards in “La Comarca Lagunera” (Gadsden et al., 2006) makes it possible to consider a conservation strategy that could include several potential flagship

species and focused to be an extensive protective umbrella of several vulnerable habitats in this area. In this way simultaneously would be protected other regional flora and fauna. Efficient conservation strategies need precise estimates of the spatial distributions of the species they are trying to protect (Johnson and Gillingham, 2005). Ecological niche modeling also known as bioclimatic modeling or climate envelope modeling is a research tool developed to produce spatially-explicit distributional hypotheses of species (Araújo and Guisan, 2006). These models have been used to successfully predict the potential distribution of species in transformed compared to untransformed habitat (Sánchez-Cordero et al., 2005). With this approach, distributional shifts caused by climatic change or habitat transformation can be estimated based on the fact that the niche model is characterized in ecological space conditions with which a species is associated (Martínez-Meyer et al., 2004). ENMs can be developed and tested based on relatively small samples of occurrences, as few as 10e20 points in some cases (Pearson et al., 2006a). In this study, we use ecological niche modeling and geographic information system (GIS) techniques to evaluate the effect of global climate change and land-cover change expected for the next 10 and 40 yr on the geographic distribution of C. antiquus and S. cyanostictus. 2. Materials and methods 2.1. Study species C. antiquus and S. cyanostictus are both endemic to the central Chihuahuan Desert, C. antiquus in southwestern Coahuila and S. cyanostictus in southwestern and southeastern Coahuila and western Nuevo León, Mexico (Lemos-Espinal and Smith, 2007). They have highly restricted geographic distributions and they are sympatric, occurring together in three small mountain ranges (Sierra Texas, Sierra Solis, and Sierra San Lorenzo) in the Mayrán Basin (Gadsden et al., 2006) (Fig. 1). C. antiquus is strongly saxicolous, it is found around rocky cliffs in brushy hills of the Chihuahua Desert where they commonly bask at certain times of the day, determine by temperature and prey

Fig. 1. Known occurrence of Crotaphytus antiquus (circles) and Sceloporus cyanostictus (triangles) in the Chihuahuan Desert. The localities were gathered directly of lizards observed in the geographic distribution of each species and from published records.

H. Gadsden et al. / Journal of Arid Environments 83 (2012) 1e9

availability. Males are territorial at least in the breading season, in spring and early summer. Dominant males have been observed perching on top of large rocks, with up to 3 females perching on quite small rocks no more than 10 m distant. Two clutches of 3e4 eggs may be deposited in one season. They are active from May to October and from 10:00 am to 6:00 pm. The species, as other members of the genus, feeds mainly on insects (Lemos-Espinal and Smith, 2007; Lemos-Espinal et al., 2012). S. cyanostictus is primarily restricted to rock walls and boulders along dry streambeds in canyons where low xeric shrubs provide partial shade as well as the insect prey on which the lizard survive. In Sierra Texas and Sierra San Lorenzo it occurs at low and middle elevations on the sides of rocky mountains, where they occupy deep crevices. In these areas they occur in the same rocky formations as C. antiquus. Males with their bright emerald-green color are conspicuous in their strong contrast with the brownish color of the rocks, which before the rain season starts are usually totally devoid of vegetation. Juveniles have been observing eating ants, but adults and juveniles in captivity accept any small insect, including mealworms (Lemos-Espinal and Smith, 2007; Lemos-Espinal et al., 2012). S. cyanostictus is primarily restricted to the heavily eroded, almost vertical strata of limestone that form the outermost formation of anticlines. Adults were found almost exclusively on basement rock outcropping and were usually associated with vertical or at least sloping surfaces. At San Lorenzo and Sierra Texas, S. cyanostictus can be seen basking on the boulders of the foothills of these Sierras, same boulders used by C. antiquus. C. antiquus seems to be a relic of an evolutionary line that led to Crotaphytus reticulatus (McGuire et al., 2007). There are several fossil records that have been assigned to Crotaphytus collaris in northern Mexico, it is likely that a large and widespread population of C. collaris occurred in most of northern Mexico, including “La Comarca Lagunera”, and during the repeated glaciations of the Pleistocene this widespread population was divided in small subpopulations, one of them limited to the Sierras San Lorenzo, Solis, and Texas of southwestern Coahuila. This isolated population evolved in rocky conditions. Subsequent to these vicariant glaciations the surrounded area of these mountain ranges was surrounded by water which was recently dried out and resulted in the actual desert characterized by flat sandy areas. A similar scenario seems likely in S. cyanostictus, where individuals from southwestern Coahuila limited to the ranges of San Lorenzo, Solis, and Texas, show a bright green coloration quite different from those from extreme eastern Coahuila and western Nuevo León, where most individuals are blue, which would account for species range. Apparently an ancestral Sceloporus jarrovii type species was widespread in northern Mexico, during the Pleistocene glaciations a numbers of subpopulations were isolated in mountain ranges of western Coahuila and in the actual Sierra Madre Oriental leading to a group of bright colored species which common ancestor is S. jarrovii (Wiens et al., 1999).

2.2. Study area The area lies between latitudes 25 310 2600 N and 26 370 0200 N and longitudes 103 150 5000 W and 101 150 5200 W in southwestern and southeastern of Coahuila. The elevation oscillates between 1109 and 1170 m above sea level. The climate of this region is seasonal, with highest temperature and rainfall occurring in summer. Mean annual precipitation is 239 mm and mean monthly temperature ranges from 13  C in January to 28  C during the summer. The vegetation was dominated by Agave lechuguilla, Fouqueria splendens, Yuca filifera, Opuntia rufida, and Opuntia rastrera and is within the Lechuguilla Scrub (Matorral Xerófilo-Rosetófilo with A. lechuguilla) plant community

3

(Rzedowski, 1978). Throughout the study site, rocks, and rock faces with crevices were numerous. 2.3. Modeling species’ distributions Several methods have been developed to model the ecological niche and predict the geographic distribution of a species (Elith et al., 2006). These approaches combine occurrence data with ecological and environmental variables to create a model of the species’ requirements for the examined variables. Primary occurrence data exist in the form of georeferenced coordinates of latitude and longitude, and digital maps (GIS raster layers) of environmental variables. The process consists of modeling the ecological requirements (ecological niche) of the species and projecting the model onto a geographic landscape in order to identify areas of potential distribution (Nix, 1986). These methods assume that the ecological niche of a species can be defined as an n-dimensional hypervolume in which each axis represents the biotic and abiotic conditions under which populations of a species can persist indefinitely without immigration (Holt and Gaines, 1992). One of the characteristics of the ecological niche is that it tends to be relatively stable over evolutionary time (niche conservatism): in general, species may change their biotic and abiotic requirements, but in a slow way (Holt and Gaines, 1992; Peterson and Cohoon, 1999). Definition of the limiting factors in the distribution of a species allows reciprocal predictions of the species distribution before and after severe climatic changes (Martínez-Meyer et al., 2004). Distributional data were represented by 52 and 24 unique localities for S. cyanostictus and C. antiquus, respectively, which were gathered directly by GPS from lizards observed during 2008e2010 in the geographic distribution of each species (Gadsden et al., Unpublished results) and from published records (Axtell and Webb, 1995; Lemos-Espinal and Smith, 2007). We used the Maximum Entropy modeling method (Philips et al., 2004, 2006), as implemented in the software MaxEnt version 3.3.1 (freely available at http://www.cs.princeton.edu/wschapire/ maxent/), to develop niche models of habitat suitability for the taxa. MaxEnt is a maximum entropy-based machine learning program that estimates the probability distribution for a species’ occurrence based on environmental constraints (Phillips et al., 2006). To assess model performance, we used Receiver Operating Characteristic (ROC) curves (Philips et al., 2004). The main advantage of ROC analysis is that the area under the ROC curve (AUC) provides a single measure of model performance, independent of any choice of threshold (Phillips et al., 2006). A perfect classifier therefore has an AUC of 1, generally, AUC values greater than 0.7 are considered to be potentially significant, while scores of 0.5 imply a predictive discrimination that is no better than random (Elith et al., 2006). Finally, to estimate which climatic variables are more relevant to determine the current geographic distribution of species in the models, a jackknife analysis was carried out using the Maximum Entropy software (MaxEnt). In this procedure each variable is excluded in turn, and a model is created with the remaining variables. Then, a model is created using each variable in isolation (Phillips et al., 2006). To conclude, the models generated in ecological space are projected onto a geographic space, MaxEnt generates a cumulative probability distribution output between 0 and 1.0%, hence, in order to create a binary map of presence/absence of the species we used the threshold value. We used the minimum cumulative value of training sample points as a threshold (Phillips et al., 2006). Using the cut off value, models were classified in to binary (1 or 0) or presence-absence model. These models generated are based solely on environmental factors and do not take into account biotic or historical factors, which may prevent species from occupying their

4

H. Gadsden et al. / Journal of Arid Environments 83 (2012) 1e9

distributional potential in full; thus produced maps should be considered potential distribution models, rather than historic/ actual distribution maps (Soberón and Peterson, 2005). Distribution models were generated using different environmental surfaces: 1) topographic data (slope and topographic index; Geological Survey of the United States; http://edcdaac.usgs.gov/ gtopo30/hydro/); and 2) environmental data obtain from Worldclim e Global Climate Data (http://www.worldclim.org) at a spatial resolution of w1 km2 (Table 1). These climate data are derived from monthly temperature and rainfall values and represent biologically meaningful variables for characterizing species distribution (Nix, 1986). The Worldclim data layers included 11 temperature and eight precipitation metrics, expressing spatial variations in annual means and extremes or limiting climatic factors. These climate surfaces are used in many applications, particularly in environmental, agricultural and biological sciences (Jones and Gladkov, 2003). The spatial resolution of the climate surfaces used in a particular study depends on the needs for that application and on the data available. For many applications, data at a fine (1 km2) spatial resolution are necessary to capture environmental variability that can be partly lost at lower resolutions (Hijmans et al., 2005). Projections of future distributions were based on the same topographic and edaphic data and estimates of the bioclimatic parameters for the next 10 and 40 yr, using the Canadian Climate Centre, CGCM2 SRES A2 scenario (Hijmans et al., 2005; http://www. worldclim.org/futdown.htm). The different scenarios depend on different future atmospheric compositions resulting from different assumptions regarding world development. The A2 scenario (liberal model) is described as “a very heterogeneous world”. The underlying theme is that strengthening regional cultural identities, high population growth, and less concern for rapid economic development. Therefore, to decide the use of the A2 scenario we assumed that in the coming decades of this century will be preserved an inertial process of a general pattern of economic and industrial development very similar to the present time and with cumulative greenhouse effect. This scenario predicts high concentrations of atmospheric CO2 and yields global increases in temperatures predicted for 2100 of 3.0e5.2  C (IPCC, 2001). This high-emissions scenario (“business as usual”) corresponds to a CO2 concentration by end of century of more than three times the preindustrial level. This scenario prediction is consistent with temperatures simulated under a broad array of climate change models. The variability in projected future temperatures across simulations using the same emissions scenarios is indicative of variability in the sensitivity of the modeled climate systems to increased greenhouse gases. It is important to note, however, that virtually all climate models project warmer springs and summers will probably occur over world in coming decades under plausible future emissions scenarios. Although it is difficult to predict global greenhouse gas emission rates far into the future, it is stressed that projections for up to 2050 show little variation between different emission scenarios, as these near-term changes in climate are Table 1 Current distribution, habitat loss caused by anthropogenic habitat conversion, and predicted future distributional area of Sceloporus cyanostictus and Crotaphytus antiquus using one climatic scenario drawn from the Canadian Climate Centre, CGCM2 SRES A2 for two periods of time, 2020 and 2050. All values are in km2. Sceloporus cyanostictus Crotaphytus antiquus Current distribution 1058 Undisturbed area 822 Predicted Potential distribution 414 CCCMA SRES A2 in 2020 Predicted Potential distribution 170 CCCMA SRES A2 in 2050

232 204 174 58

strongly affected by greenhouse gases that have already been emitted and will stay in the atmosphere for the next 50 years (IPCC, 2007). All maps for the present and future were resample to a spatial resolution of 30 arc seconds; this is equivalent to about 0.86 km2 at the equator and less elsewhere and commonly referred to w1 km2 resolution (Hijmans et al., 2005). In fact, if we superimposing a grid of 1  1 km cells onto the species distributional data and counting the number of cells occupied by each species is the same number of km2. Several authors have used the same method at least in latitudes which includes Mexico (Parra-Olea et al., 2005; Téllez-Valdés and Dávila-Aranda, 2003). After models were produced, we used a digital map from the Carta de Uso del Suelo y Vegetación, 1:250 000, Serie III (INEGI, 2005) as the base for current land-use and vegetation. Habitats transformed into agro systems and rural or urban settlements were eliminated from current and future distribution models because we considered that these constitute unsuitable habitat for the species, and we assumed that they will not be retransformed to undisturbed conditions in the future. In fact we have explored these areas and have found no individuals of S. cyanostictus and C. antiquus. In contrast to studies of other species (Parra-Olea et al., 2005), our projections assumed an inability of either species to disperse outside their current range because these lizards are specialized and restricted exclusively to rocky areas, and their vagility is probably very low. Thus projections assume that species would inhabit only those portions of their present distributional areas that remain habitable. 3. Results Although sample sizes for both species were small, models produced were acceptably accurate. All results of the ENM generated for each species (Table 1, Fig. 2) had AUC values >0.9, implying potentially significant results. This predictive ability is observed in the zero omission error registered in the models for both species, as well as the fact that the areas obtained for the current distribution do not show high overprediction, as compared to the known distribution. Estimated potential distribution area for C. antiquus is 232 km2; however, this area is reduced when considering the effect of human activities, mainly, mining, residential developments, and human settlement; according to the land-use map, 12% of the area has been impacted (Table 1, Fig. 2a). In comparison, the current potential distribution of S. cyanostictus is 1058 km2; this area is reduced 22% when areas that have been transformed are removed (Table 1, Fig. 2b). The MaxEnt model’s internal jackknife test of variable importance showed that Annual Precipitation, Precipitation of Wettest Month, Precipitation of Driest Month, Precipitation of Wettest Quarter, Precipitation of Driest Quarter, Precipitation of Warmest Quarter, and Precipitation of Coldest Quarter were the most important variables predictors of C. antiquus’ habitat distribution, and with these same variables was also important for C. antiquus the Minimum Temperature of Coldest Month (Table 2). These variables presented the higher training gain (that is, contained most information) compared to other variables (Fig. 3a and b). Strong climatic changes are projected to occur in the central Chihuahuan Desert in the coming decades, particularly in the period 2020e2050 (Table 2). Effects of these climatic changes are expected to affect severely the geographic distribution of both species. In general, under the A2 scenario a 25% reduction of the modeled range of C. antiquus and 61% of S. cyanostictus is expected by 2020. In 2050 the picture looks even inferior since the range of C. antiquus will be reduced 75% and the distributional area of S. cyanostictus is predicted to be decreased 84% (Table 1, Fig. 2a and b).

H. Gadsden et al. / Journal of Arid Environments 83 (2012) 1e9

5

Fig. 2. Potential distributions models of (a) Crotaphytus antiquus and (b) Sceloporus cyanostictus for liberal climate change scenario (SRES A2) for 2020, and 2050. Color correspond to light-gray ¼ distributional areas lost by habitat conversion, dark-gray ¼ current potential distribution, and black ¼ predicted distribution remaining in future. Gray lines area indicates highways and roads, black lines are the political boundaries of Durango and Coahuila, and black stars are main cities. The maps correspond to a zooming of the study area.

4. Discussion This work is one of the first efforts to evaluate the possible future consequences of two main drivers of current global change, habitat destruction and climate change, on the distribution of endemic reptiles in Mexico. In the case of S. cyanostictus and C. antiquus, their limited distribution in small areas of rocky hills, in addition to their low vagility, reduces the likelihood of migration to sites where these saxicolous conditions can be maintained (Gadsden, Unpublished results). The low vegetative diversity and cover in these rocky ecosystems probably makes them particularly vulnerable to the multiple alterations to which they are being subjected.

Alteration of the habitat of S. cyanostictus and C. antiquus by urbanization, agricultural use, and cattle ranching is severe. The spatial analysis used in our study found that the major portion of suitable habitat for these species is located in the middle of “La Comarca Lagunera” area, one of the most important textile, agricultural, and industrial regions in northern Mexico, bordered by large human settlements, such as Torreón, Gómez Palacio and Lerdo. In addition to the high human density in the area, roads also contribute to the problem. Highways and roads are major contributors to habitat fragmentation because they divide continuous landscapes into smaller patches and convert interior habitat into edge habitat (Noss and Cooperrider, 1994).

6

H. Gadsden et al. / Journal of Arid Environments 83 (2012) 1e9

Table 2 Mean and standard deviation values of the bioclimatic parameters that determine the distribution of Sceloporus cyanostictus (top of each cell) and Crotaphytus antiquus (bottom of each cell) respectively. The values shown are for the present, 2020 and 2050, using the liberal climatic scenario (CGCM2 SRES A2). * Values of the main bioclimatic parameters according to the jackknife analysis. Bioclimatic parameters bio1 ¼ Annual Mean Temperature ( C) bio2 ¼ Mean Diurnal Range (Mean of monthly (max tempmin temp)) ( C) bio3 ¼ Isothermality (bio2/bio7)  100 bio4 ¼ Temperature Seasonality (SD  100) bio5 ¼ Maximum Temperature of Warmest Month ( C) bio6 ¼ Minimum Temperature of Coldest Month ( C) bio7 ¼ Temperature Annual Range (bio5-bio6) ( C) bio8 ¼ Mean Temperature of Wettest Quarter ( C) bio9 ¼ Mean Temperature of Driest Quarter ( C) bio10 ¼ Mean Temperature of Warmest Quarter ( C) bio11 ¼ Mean Temperature of Coldest Quarter ( C) *bio12 ¼ Annual Precipitation (mm) *bio13 ¼ Precipitation of Wettest Month (mm) *bio14 ¼ Precipitation of Driest Month (mm) bio15 ¼ Precipitation Seasonality (Coefficient of Variation) *bio16 ¼ Precipitation of Wettest Quarter (mm) *bio17 ¼ Precipitation of Driest Quarter (mm) *bio18 ¼ Precipitation of Warmest Quarter (mm) *bio19 ¼ Precipitation of Coldest Quarter (mm) Slope (0e90 ) Topographic Index (0e360 )

Present

2020 year

2050 year

20.8 21.0 16.9 17.0

   

0.4 0.3 0.4 0.2

21.4 22.2 17.6 17.9

   

0.5 0.2 0.3 0.2

22.3 23.1 17.8 18.0

   

0.4 0.2 0.3 0.2

53.4 53.8 516.5 519.0 35.6 35.8 4.3 4.4 31.3 31.4 25.8 26.0 18.6 18.9 26.6 26.8 13.7 13.9 212.5 207.3 37.9 36.7 2.9 2.5 68.3 66.4 105.6 102.6 12.9 11.9 97.3 94.8 24.2 25.1 36.8 23.4 52.3 38.7

                                     

0.5 0.6 6.2 9.6 0.7 0.2 0.1 0.1 0.6 0.2 0.4 0.4 0.8 0.3 0.4 0.4 0.3 0.2 15.6 11.5 3.5 2.6 0.7 0.5 1.6 2.3 8.7 7.0 1.8 1.0 6.3 6.7 0.8 0.9 22.2 12.0 54.3 10.8

54.0 54.0 525.7 543.8 36.6 37.7 4.3 4.7 32.3 33.0 26.3 27.2 18.9 19.7 27.2 28.1 14.0 14.5 201.5 183.1 48.0 42.3 2.1 1.2 78.3 73.7 102.5 89.2 11.0 9.3 83.5 73.9 19.4 19.6

                                 

0.9 0.5 13.7 7.4 0.6 0.3 0.2 0.1 0.4 0.2 0.7 0.6 0.7 0.2 0.6 0.3 0.3 0.1 14.9 8.6 4.3 2.9 0.7 0.4 2.1 2.8 10.2 5.3 1.2 0.6 7.2 4.1 0.9 1.0

54.7 54.0 522.3 540.1 37.6 38.7 5.3 5.7 32.3 33.0 26.9 26.8 17.7 20.1 28.0 29.0 14.9 15.5 191.2 172.7 43.9 38.5 2.8 2.0 73.9 69.3 93.9 82.0 12.7 10.3 75.4 65.9 19.2 18.6

                                 

0.8 0.5 13.1 7.9 0.6 0.3 0.2 0.1 0.4 0.2 0.4 0.4 1.8 1.2 0.6 0.3 0.3 0.1 14.4 7.8 4.0 2.6 0.4 0.2 2.8 2.6 9.1 4.8 1.3 0.8 6.7 3.8 1.0 1.1

According to our results, habitat transformation has resulted in the loss of 12% of the range of C. antiquus and 22% of the range of S. cyanostictus. Unfortunately, there is no formal protection of any region within their current ranges. In addition, expectations of climatic changes in the region suggest that 25% for C. antiquus and 16% for S. cyanostictus of the remaining current range will continue to be habitable by 2050. This is mainly the result of a predicted drastic drop in rainfall levels during the summer and a global temperature rise in the period 2020e2050. These parameters were the main driving factors determining the distribution of these species, according to the jackknife analysis (Table 2). This could have correspondingly dramatic effects on population dynamics and the intensity and timing of reproduction in annual lizard species (as C. antiquus and S. cyanostictus) inhabiting isolated rocky hills in the Chihuahuan Desert. In general, possible responses of species to climate change include niche tracking and adaptation (Holt, 1990). When species are vagile enough, individuals are able to move relatively long distances in search for suitable areas. Alternatively, if species are capable of rapid evolutionary change, or have a wide range of

physiological tolerances, adjustments to changing conditions may be possible. Failing both, extinction is the likely result (Holt, 1990). Unfortunately, the current warming event is causing highly accelerated climatic changes (IPCC, 2007). According to Huey et al. (2003), lizards cannot evolve rapidly enough to track current climate change because of constraints arising from the genetic architecture of thermal preference. A phylogenetic correlation between body temperature (Tb) and maximum critical temperature (CTmax) constraints adaptation (Sinervo et al., 2010). In the same way the evolution response necessary to keep pace with climate change is further constrained by low heritability for Tb (Sinervo et al., 2010). Coupled with the fact that S. cyanostictus and C. antiquus have reduced vagility and low population sizes (Gadsden et al., Unpublished results), both species appear to be facing a critical situation in the near future. This has been observed for several other reptile species elsewhere (Ballesteros-Barrera et al., 2007). For example, since 1987, species of Anolis lizards that live in the cloud forest of Monteverde, Costa Rica have disappeared due to an increase of temperature and reduction in humidity (Schneider, 1999). Our results indicate that S. cyanostictus and C. antiquus face a critical situation. Both species are currently at a higher risk due to habitat transformation and global climate change. This merits a serious and critical review for formal protection of the three small mountain ranges (Sierra Texas, Sierra Solis and Sierra San Lorenzo) in this area, and a change in conservation status to “critically endangered” for both species. Recent field studies suggested that ecotourism, the allocation of critical areas for conservation and research that consider the potential effects of climate change, the active participation of local people, and a strong communication campaign could contribute to the conservation of these species. Our study area is located within the “Comarca Lagunera” which is entirely contained in the Mapimian subdivision of the Chihuahuan Desert Ecoregion (Morafka et al., 1992), so ecoregional complexity fails to account for the lizard species richness found there. La Laguna depositional sink includes several isolated inselbergs of Cretaceous limestone protruding through a “sea” of aeolian sands and lake-bottom sediments (Lehmann et al., 1999). The isolation of these mountains has apparently fostered speciation in saxicolous lizards as evidenced by the seven endemic rocky-habitat lizards in this area. However, for insulated species with restricted habitat requirements, global climate change, habitat modifications or disturbance can be devastating. Presently, anthropogenic degradation of the habitats of these and other species in the area is accelerating. In addition to the endemic sand dune lizard Uma exsul, The “Comarca Lagunera” basin contains several endemic saxicolous lizard species, including C. antiquus and S. cyanostictus (Morafka et al., 1992; Gadsden et al., 2006). However, despite of the important endemism of “La Laguna” only the Coahuila fringe-toed lizard U. exsul is seen potentially as a “flagship” species and has received protection under federal endangered species legislation in Mexico. Nevertheless, recently both S. cyanostictus and C. antiquus were categorized by IUCN (Red List of Threatened Species) as endangered species (Hammerson et al., 2007; Vazquez-Díaz et al., 2007). Therefore the high endemism of lizards in “La Laguna” (Gadsden et al., 2006) makes it possible to consider a holistic conservation strategy that could include a greater number of potential flagship species for the purpose of having a more extensive protective umbrella of several threatened habitats of this basin. In this way also would be protected further biodiversity that includes other flora and fauna of the region in various habitats impacted by human actions. Finally, a word of caution concerning our results is appropriate here. Different sources of uncertainty may be affecting our estimations. While ecological niche modeling predicts potential

H. Gadsden et al. / Journal of Arid Environments 83 (2012) 1e9

7

Fig. 3. Results of jackknife evaluations of relative importance of predictor variables for (a) Crotaphytus antiquus and (b) Sceloporus cyanostictus Maxent models. The main bioclimatic variables for both species were: bio12 ¼ Annual Precipitation, bio13 ¼ Precipitation of Wettest Month, bio16 ¼ Precipitation of Wettest Quarter, bio17 ¼ Precipitation of Driest Quarter, bio18 ¼ Precipitation of Warmest Quarter, bio19 ¼ Precipitation of Coldest Quarter, and bio14 ¼ Precipitation of Driest Month. The Minimum Temperature of Coldest Month (bio6) was only relevant to C. antiquus.

8

H. Gadsden et al. / Journal of Arid Environments 83 (2012) 1e9

geographic distributions of species, certain areas may not be occupied currently, due to factors external to the model, such as historical constraints, species interactions, geographic barriers, and changes in land-use patterns (Sánchez-Cordero et al., 2005). In this case, the modeled distribution area may be overestimating the actual distributional range of species, since both lizards inhabit highly specific saxicolous environments within the area. For example, recent field studies for S. cyanostictus and C. antiquus estimated around 133 km2 of remaining rocky hills habitat (Gadsden, Unpublished results). Additionally, future climate change scenarios include a great deal of uncertainty (Murphy et al., 2004), and ecological niche modeling algorithms involve some level of uncertainty that is exacerbated in projections to simulated scenarios (Pearson et al., 2006b). In addition, Araújo and New (2007) suggest that using ensemble forecasting approaches has a clear advantage over single model forecasts. These approaches fit in to a number of alternative models and explore the range of resulting projections. If used suitably, either individually, in combination, or in hybrid form, these approaches can enable more robust decision making in the face of uncertainty, and have much to offer to conservation planning. Likewise, Buisson et al. (2010) and Carvalho et al. (2010) propose that our dramatic predictions should be interpreted with prudence given all the uncertainties in the process and recommend paying more consideration to the following aspects when evaluating the impacts of climate change on biodiversity: (a) using several statistical methods in an ensemble forecasting framework, (b) using climate projections from diverse General Circulation Models (GCMs) to evaluate impacts in the short and longterm, (c) using several climate scenarios only for impacts in the longterm, (d) the scale of the analysis, (e) always providing maps of uncertainty in combination with maps of projected impacts, (f) taking into account species biological attributes when combining responses of individual species to evaluate impacts on assemblages, and (g) specific dispersal capacities. Following these recommendation efforts could be made to achieve a more realistic understanding of the future impacts of climate change on biodiversity, allowing management and conservation decisions to be taken with awareness of the intrinsic uncertainty in those impacts. Nevertheless, we consider our results as a particular approach of an “early warning” of a possible outcome if current worst-case land-use transformation and climatic trends continue. Acknowledgments This study was supported by a SEP-CONACYT grant (43142-Q). References Araújo, M.B., Guisan, A., 2006. Five (or so) challenges for species distribution modeling. Journal of Biogeography 33, 1677e1688. Araújo, M., New, M., 2007. Ensemble forecasting of species distributions. Trends in Ecology and Evolution 22, 42e47. Axtell, R.W., Webb, R.G., 1995. The new Crotaphytus from southern Coahuila and adjacent states of East-central Mexico. Bulletin of the Chicago Academy of Sciences 16, 1e15. Ballesteros-Barrera, C., Martínez-Meyer, E., Gadsden, H., 2007. Effects of land- cover transformation and climate change on the distribution of two microendemic lizards, genus Uma, of northern Mexico. Journal of Herpetology 41, 732e739. Buisson, L., Thuiller, W., Casajus, N., Lek, S., Grenouillet, G., 2010. Uncertainty in ensemble forecasting of species distribution. Global Change Biology 16, 1145e1157. Carvalho, S., Brito, J., Crespo, E., Possingham, H., 2010. From climate change predictions to actions e conserving vulnerable animal groups in hotspots at a regional scale. Global Change Biology 16, 3257e3270. de Chazal, J., Rounsevell, M.D.A., 2009. Land-use and climate change within assessments of biodiversity change: a review. Global Environmental Change 19, 306e315. Elith, J., Graham, C.H., Anderson, R.P., Dudík, M., Ferrier, S., Guisan, A., Hijmans, R.J., Huettmann, F., Leathwick, J.R., Lehmann, A., Li, J., Lohmann, L.G., Loiselle, B.A.,

Manion, G., Moritz, C., Nakamura, M., Nakazawa, Y., Overton, J.M., Peterson, A.T., Phillips, S.J., Richardson, K., Scachetti-Pereira, R., Schapire, R.E., Soberón, J., Williams, S., Wisz, M.S., Zimmermann, N.E., 2006. Novel methods improve prediction of species’ distributions from occurrence data. Ecography 29, 129e151. FAO (Organization for Food and Agriculture of the United Nations), 2005. Global Forest Resources Assessment. In: Progress towards Sustainable Forest Management, Rome 2006, 352 pp. ISSN 0258-6150, FAO-Bookinfo. Flores-Villela, O., Canseco-Márquez, L., 2004. Nuevas especies y cambios taxonómicos para la herpetofauna de México. Acta Zoológica Mexicana 20, 115e144. Gadsden, H., Estrada-Rodríguez, J.L., Leyva-Pacheco, S.V., 2006. Checklist of amphibians and reptiles of the “Comarca Lagunera” in Durango-Coahuila, Mexico. Bulletin of the Chicago Herpetological Society 41, 2e9. Godfrey, M.H., D’Amato, A.F., Marcovaldi, M.A., Mrosovsky, N., 1999. Pivotal temperature and predicted sex ratios for hatchling hawksbill turtles from Brazil. Canadian Journal of Zoology 77, 1465e1473. Hammerson, G.A., Vazquez-Díaz, J., Gadsden, H., Quintero-Díaz, G.E., PonceCampos, P., Lavin, P., 2007. Crotaphytus antiquus. In: IUCN 2011. IUCN Red List of Threatened Species. Version 2011.1. www.iucnredlist.org. Hijmans, R.J., Cameron, S.E., Parra, J.L., Jones, P.G., Jarvis, A., 2005. Very high resolution interpolated climate surfaces for global land areas. International Journal of Climatology 25, 1965e1978. Holt, R.D., Gaines, M.S., 1992. The analysis of adaptation in heterogeneous landscapes: implications for the evolution of fundamental niches. Evolutionary Ecology 6, 433e447. Holt, R.D., 1990. The microevolutionary consequences of climate change. Trends in Ecology and Evolution 5, 311e315. Huey, R.B., Hertz, P.E., Sinervo, B., 2003. Behavioral drive versus behavioral inertia in evolution: a null model approach. American Naturalist 161, 357e366. INEGI (Instituto Nacional de Estadística, Geografía e Informática), 2005. Carta de Uso del Suelo y Vegetación. In: 1:250 000. Serie III. México. IPCC (Intergovernmental Panel on Climate Change), 1995. IPCC Second Assessment. In: Climate Change 1995. IPCC Secretariat, Geneva, Switzerland. IPCC (Intergovernmental Panel on Climate Change), 2001. Climate Change 2001: Impacts, Adaptation and Vulnerability. Cambridge University Press. IPCC (Intergovernmental Panel on Climate Change), 2007. Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. IPCC Secretariat, Geneva, Switzerland. Johnson, C.J., Gillingham, M.P., 2005. An evaluation of mapped species distribution models used for conservation planning. Environmental Conservation 32, 1e12. Jones, P.G., Gladkov, A., 2003. FloraMap. In: A Computer Tool for Predicting the Distribution of Plants and Other Organisms in the Wild. Version 1.02. Centro Internacional de Agricultura Tropical, Cali, Colombia. Lehmann, C., Osleger, D.A., Montanez, I.P., Sliter, W.V., Arnaud-Vanneau, A., Banner, J.L., 1999. Evolution of Cupido and Coahuila carbonate platforms, early Cretaceous, northeastern Mexico. Geological Society of America Bulletin 111, 1010e1029. Lemos-Espinal, J.A., Smith, H.M., 2007. Anfibios y Reptiles del estado de Coahuila, México/Amphibians and Reptiles of the state of Coahuila, México. CONABIO, México. Lemos-Espinal, J.A., Smith, H.M., Dixon, J.R., Cruz, A., 2012. Anfibios y Reptiles de Sonora, Chihuahua y Coahuila, México / Amphibians and Reptiles of Sonora, Chihuahua and Coahuila. Volumen I. CONABIO, i-x þ 1-600; Volumen II CONABIO, i-x ¼ 1-520, México. Martínez-Meyer, E., Peterson, A.T., Hargrove, W.W., 2004. Ecological niches as stable distributional constraints on mammal species, with implications for Pleistocene extinctions and climate change projections for biodiversity. Global Ecology and Biogeography 13, 305e314. McGuire, J.A., Linkem, C.W., Koo, M.S., Hutchison, D.W., Lappin, A., Orange, D.I., Jaeger, J.R., 2007. Mitochondrial introgression and incomplete lineage sorting through space and time: phylogenetics of crotaphytid lizards. Evolution 61, 2879e2897. Mittermeier, R., Goettsch, C., 1992. La importancia de la diversidad biológica de México. In: Sarukhán, J., Dirzo, R. (Eds.), México ante los retos de la biodiversidad. Conabio, México, D.F, pp. 57e62. Morafka, D.J., Adest, G.A., Reyes, L.M., Aguirre, G.L., Lieberman, S.S., 1992. Differentiation of North American Deserts: a phylogenetic evaluation of a vicariance model. In: Biogeography of MesoAmerica: Proceedings of a Symposium, Mérida, Yucatán, México. October 1984. Tulane Studies in Zoology and Botany, Supplementary Publication Number 1, pp. 195e226. Murphy, J.M., Sexton, D.M., Barnett, D.N., Jones, G.S., Webb, M.J., Collins, M., Stainforth, D.A., 2004. Quantification of modelling uncertainties in a large ensemble of climate change simulations. Nature 430, 768e772. Nix, H.A., 1986. A biogeographical análisis of the Australian elapid snakes. In: Longmore, R. (Ed.), Atlas of Elapid Snakes. Australian Flora and Fauna Series, pp. 5e15. Noss, R.F., Cooperrider, A.Y., 1994. Saving Nature’s Legacy: Protecting and Restoring Biodiversity. Defenders of Wildlife and Island Press, Washington, D.C. Parra-Olea, G., Martínez-Meyer, E., Pérez-Ponce de León, G., 2005. Forecasting climate change effects on salamander distribution in the highlands of central Mexico. Biotropica 37, 202e208. Pearson, R.G., Raxworthy, C.J., Nakamura, M., Peterson, A.T., 2006a. Predicting species distributions from small numbers of occurrence records: a test case using cryptic geckos in Madagascar. Journal of Biogeography 34, 102e117.

H. Gadsden et al. / Journal of Arid Environments 83 (2012) 1e9 Pearson, R.G., Thuiller, W., Araújo, M.B., Brotons, L., Martínez-Meyer, E., McClean, C., Miles, L., Segurado, P., Dawson, T.P., Lees, D., 2006b. Modelbased uncertainty in species’ range prediction. Journal of Biogeography 33, 1704e1711. Peterson, A.T., Cohoon, K.C., 1999. Sensitivity of distributional prediction algorithms to geographic data completeness. Ecological Modelling 117, 159e164. Philips, S.J., Dudik, M., Schapire, R.E., 2004. A maximum entropy approach to species distribution modeling. In: Proceedings of the 21st International Conference on Machine Learning. ACM Press, New York, pp. 655e662. Phillips, S.J., Anderson, R.P., Schapire, R.E., 2006. Maximum entropy modeling of species geographic distributions. Ecological Modelling 190, 231e259. Pounds, J.A., Fogden, M.P.L., Campbell, J.H., 1999. Biological response to climate change on a tropical mountain. Nature 398, 611e615. Price, M., Harrison, C.R., Lazcano-Villarreal, D., 2010. Sceloporus cyanostictus. Mexico: Nuevo León. Herpetological Review 41, 108. Rzedowski, J., 1978. In: Vegetación de México. Limusa, México. Sánchez-Cordero, V., Illoldi-Rangel, P., Linaje, M., Sarkar, S., Peterson, A.T., 2005. Deforestation and Extant distributions of Mexican endemic Mammals. Biological Conservation 126, 465e473. Santos-Barrera, G., Pacheco, J., Ceballos, G., 2004. Áreas prioritarias para la conservación de los reptiles y anfibios de México. Biodiversitas 57, 1e6. Schneider, S.H., 1999. Amphibian Declines in the Cloud Forest of Costa Rica: Responses to Climate Change?. USGCRP Seminar, 29 September 1999.

9

Sinervo, B., Mendez-De la Cruz, F., Miles, D.B., Heulin, B., Bastiaans, E., VillagranSanta Cruz, M., Lara-Resendiz, R., Martínez-Méndez, N., CalderonEspinosa, M.L., Mesa-Lázaro, R.N., Gadsden, H., Avila, L.J., Morando, M., De la Riva, I.J., Victoriano Sepulveda, P., Duarte Rocha, C.F., Ibargüengoyta, N., Aguilar Puntriano, C., Massot, M., Lepetz, V., Oksanen, T.A., Chapple, D., Bauer, A., Branch, W., Clobert, J., Sites Jr., J.W., 2010. Erosion of lizard diversity by climate change and Altered thermal niches. Science 328, 894e899. Soberón, J., Peterson, A.T., 2005. Interpretation of models of fundamental ecological niches and species’ distributional areas. Biodiversity Informatics 2, 1e10. Téllez-Valdés, O., Dávila-Aranda, P., 2003. Protected areas and climate change: a case study of the cacti in the Tahucan-Cuicatlán Biosphere Reserve, México. Conservation Biology 17, 846e853. Thomas, Ch, Cameron, A., Green, R.E., Bakkenes, M., Beaumont, L.J., Collingham, Y.C., Erasmus, B.F.N., de Siqueira, M.F., Grainger, A., Hannah, L., Hughes, L., Huntley, B., van Jaarsveld, A.S., Midgley, G.F., Miles, L., Ortega-Huerta, M.A., Towsend Peterson, A., Phillips, O.L., Williams, S.E., 2004. Extinction risk from climate change. Nature 427, 145e148. Vazquez-Díaz, J., Gadsden, H., Quintero-Díaz, G.E., Ponce-Campos, P., Lavin, P., 2007. Sceloporus cyanostictus. In: IUCN 2011. IUCN Red List of Threatened Species. Version 2011.1. www.iucnredlist.org. Wiens, J.J., Reeder, T.W., Nieto Montes de Oca, A., 1999. Molecular phylogenetics and evolution of sexual dichromatism among populations of the Yarrow’spiny lizard (Sceloporus jarrovii). Evolution 53, 1884e1897.

Loading...

Effects of land-cover transformation and climate change - FCB-UJED

Journal of Arid Environments 83 (2012) 1e9 Contents lists available at SciVerse ScienceDirect Journal of Arid Environments journal homepage: www.els...

793KB Sizes 0 Downloads 0 Views

Recommend Documents

No documents