Annual chronology and climate response in Abies guatemalensis Rehder (Pinaceae) in Central America

Abstract

The continued expansion of dendroclimatology into Mesoamerica requires the identification and evaluation of species whose rings can be precisely dated and then statistically compared with precipitation and temperature variability in order to make inferences about past climate. Here, we establish the basis for using Abies guatemalensis Rehder (Pinaceae) for climate reconstruction in Central America. Annual crossdating in this montane species is demonstrated at high-elevation sites in the Sierra de los Cuchumatanes in western Guatemala. We find that ring width is most strongly influenced by early growing season moisture conditions, controlled by late dry season rainfall, and negatively correlated with growing season temperature. Our chronology is also significantly negatively correlated with eastern tropical Pacific sea surface temperature anomalies. Our confirmation of annual chronology and the identification of a climatic signal in this species now allow its use in local and regional paleoclimate reconstructions, as well as ecological studies.

Keywords

Central America climate dendrochronology drought Guatemala Mesoamerica tree-ring

Introduction

While the signature of anthropogenic climatic change is frequently associated with observed and ongoing increases in global mean and regional temperatures, it is concomitant changes in precipitation and the frequency and duration of drought and flooding that will have the most direct and most immediate consequences for human populations (Milly et al., 2008). Changes in regional hydroclimate will exacerbate already existing and rapidly emerging threats to sustainable water supplies from growing populations, pollution, declining infrastructure, and transboundary resource conflicts (Vorosmarty et al., 2000). Critical to mitigating the worst consequences of changes in water availability is a long-term perspective on the potential range of variability in precipitation, and an eventual integration of this knowledge within water planning and natural resources policy (e.g. Woodhouse and Lukas, 2006).

Long-term instrumental climate records are sparse in Central America. This is particularly true for high-elevation regions, which are likely to demonstrate the earliest and more severe local consequences of global climate change (Bradley et al., 2004). The limited length of instrumental records, both from meteorological stations and from satellite-based instruments, unfortunately impedes a complete understanding of the potential range of low-frequency variability of the climate system. As a result, there is considerable uncertainty in the detection and attribution of the potential atmospheric signature of anthropogenic climate change over the last several decades.

In the absence of long records, climate proxies such as tree-ring widths can be used to estimate past precipitation or temperature variability. Tree rings have been used to identify past ‘megadroughts’ with societal repercussions (Buckley et al., 2010; Stahle et al., 1998b, 2000b), provide improved understanding of decadal-scale variability (e.g. Ault and St George, 2010; Pederson et al., 2004), and place current and future hydroclimatic variability in a long-term context (Cook et al., 2010a; Woodhouse et al., 2006). Tree-ring reconstructions also provide information for resource management and planning (Barnett and Pierce, 2009; Woodhouse and Lukas, 2006) and probabilistic climate prediction (Bell et al., 2011). Networks of tree-ring chronologies have been successfully used to reconstruct spatiotemporal drought conditions during the last millennium in several regions including Monsoon Asia (Cook et al., 2010b), North Africa (Touchan et al., 2008, 2010), and North America (Cook et al., 1999, 2004). For the latter, however, the lack of tree-ring chronologies in Mesoamerica outside of Mexico thus far restricts its southern extent.

Interest in dendrochronology in Central America goes at least as far back as Schulman (1944), and some of the earliest chronology development in Mexico was done by Scott (1966), who worked with Pinus ponderosa at the Casas Grandes archaeological site in the state of Chihuahua. Over the last two decades there has been a rapid expansion of research on dendrochronology throughout Mexico (e.g. Acuna-Soto et al., 2002; Biondi, 2001; Biondi et al., 2005; Cleaveland et al., 2003; Diaz et al., 2002; Stahle et al., 1998a, 2000a, 2007, 2011a, 2011b; Therrell et al., 2002, 2004; Villanueva-Diaz et al., 2007). However, the distribution of many of the species used for dendrochronology in Mexico – including the drought-sensitive Pseudotsuga menziesii (e.g. Pohl et al., 2003; Stahle et al., 2000a; Therrell et al., 2006; Villanueva-Diaz et al., 2007) – does not extend further into Central America. While the range of Montezuma Cypress (Taxodium mucronatum) includes the river valleys of westernmost Guatemala (Stahle et al, 2011b), the continued expansion of traditional tree-ring width dendrochronology into Central America will require the evaluation of additional new species and new sites along the cordillera. Guatemala has several native montane species that hold considerable promise for use in dendrochronology, including Pinus hartwegii, Pinus ayacahuite, and Abies guatemalensis, amongst others. Szejner (2011) has recently shown that crossdating is possible in Pinus oocarpa from eastern Guatemala.

Expanding the frontier of dendrochronology further still into the neotropics presents several recognized challenges. South of Mexico it will necessarily involve the use of little-investigated species. While Evans and Schrag (2004), Anchukaitis and Evans (2010), and Anchukaitis et al. (2008) have used stable oxygen isotopes to establish chronology and estimate climate variability in trees without rings in Costa Rica, traditional ring width or density measurements require the formation of reliably annual rings that can be dated to their exact year of formation. Establishing crossdating is a prerequisite to performing climatic analysis and estimating past precipitation or temperature anomalies from tree-ring proxies. Here, we demonstrate that crossdating and the development of an annual, replicated chronology is possible from high-elevation populations of Abies guatemalensis (locally known as Pinabete or Parchac, Aguirre and de Pöll, 2007) in Guatemala. Through comparison with local meteorological data and regional-scale gridded data synthesis products, we also evaluate the climate signal recorded by the rings of this species. This investigation establishes the basis for climate reconstructions from this species in Central America.

Materials and methods

Our approach here is similar to that of Speer et al. (2004) and February and Stock (1998), and follows the strategy described by Stahle (1999) for developing tree-ring chronologies from tropical species. Stahle (1999) observes that robust crossdating combined with a significant and biophysically plausible relationship to climate can be used to support and establish the existence of annual rings in tropical species. Targeting coniferous tropical species from taxa known to develop reliably annual rings has been successful in both Central America (Acuna-Soto et al., 2002; Diaz et al., 2002; Stahle et al., 2000a) as well as southeast Asia (e.g. Buckley et al., 2007, 2010) in identifying potentially old and climate-sensitive trees with reliable crossdating.

In the mountains of western Guatemala, intact high-elevation fir community forests above 2750 m are dominated by A. guatemalensis, existing in some locations as a codominant with P. ayacahuite (Islebe et al., 1995; Veblen, 1978). Although A. guatemalensis has not previously been used for dendrochronology, the genus Abies has been widely used by tree-ring scientists in temperate latitudes, and a tropical species of Abies has also been used by Stahle and colleagues to develop a chronology at Mesa de Campanero in Mexico (ITRDB #MEXI032). Huante et al. (1991) also established crossdating in Abies religiosa in Michoacan in Mexico. Our study sites are in the Sierra de los Cuchumatanes (Figure 1), a high limestone plateau with a maximum elevation of 3828 m and evidence of Quaternary (late Wisconsin) glaciation (Roy and Lachniet, 2010). The region has a strong dry season between November and April, wet season peak rainfall in June and September, and the mid-summer ‘drought’ (Magaña et al., 1999) typical of the region (Figure 2). At all three of our sites, A. guatemalensis is the dominant tree species, occurring often in nearly pure stands on slopes and in drainages immediately downslope from the higher, drier ridgelines dominated by Pinus hartwegii. Andersen et al. (2006) observed that A. guatemalensis was typically found in cooler, moist sites, including humid valleys, and inferred that the altitudinal range of the species was restricted by precipitation as opposed to temperature. A. guatemalensis is currently classified as endangered, primarily due to overharvesting (Andersen et al., 2006; Kollmann et al., 2008).

Figure 1.

Study location in the Cuchumatanes of western Guatemala. Regional topographic features and political borders are shown. Sampling locations are indicated by red circles and a corresponding three-letter site code (see text for details). The town of Todos Santos Cuchumatán (15.51°N, 91.60°W, 2480 m) is located between and approximately equidistant from the three sampling locations. The location of Guatemala City is indicated by the black star.

Figure 2.

Mean climatology for Todos Santos Cuchumatán (15.51°N, 91.60°W, 2480 m). Precipitation mean is based on the period 1983 to 2010. Temperature mean is based on 1990 to 2010. The coefficient of variation for precipitation is calculated as the standard deviation divided by the mean and reflects the relative magnitude of the interannual variability. All data from the Guatemalan Instituto Nacional de Sismologia, Vulcanología, Meteorología e Hidrologia (INSIVUMEH; http://www.insivumeh.gob.gt/meteorologia.html) (colour figure available online).

We extracted increment cores from A. guatemalensis at three sites (Figure 1) in the Cuchumatanes following standard procedures (Stokes and Smiley, 1968). Bosque del Rancho (BDR, 15.39°N, 91.54°W, ~3500 m) is immediately downslope from La Torre, the highest point in the Cuchumatanes. Puerta al Cielo (PAC, 15.55°N, 91.58°W, ~3400 m) is approximately 10 km to the north and west of BDR at nearly the same elevation. Montaña San Juan (MSJ, 15.47°N, 91.60, ~3100 m) is a lower-elevation site to the south of the town of Todos Santos Cuchumatán and approximately 10 km west of BDR and to the south of PAC. We selected a subset of 60 cores from 30 trees in order to assess dating and climate signal in this species. Cores were dried, mounted, and sanded before being crossdated using the procedure described by Yamaguchi (1991). Cores were then measured to 0.001 mm precision and crossdating was verified using the computer program COFECHA (Holmes, 1983). The individual series were detrended using a smoothing spline with a 50% amplitude response at 67% of the series length and the master chronology was then developed as the biweight robust mean of the detrended series (Cook, 1985; Cook and Peters, 1981, 1997). Here, we use the prewhitened residual chronology in our climate analysis in order to minimize the statistical influence of autocorrelation.

We compare our chronology to local and regional gridded climate data. Local meteorological data from the region is sparse, but is available monthly (with some missing data) from the town of Todos Santos Cuchumatán back to 1983 for precipitation and 1990 for temperature (see Figure 2, http://www.insivumeh.gob.gt/meteorologia.html). The town is approximately equidistant from our three sampling locations (15.51°N, 91.60°W, 2480 m, Figure 1). For gridded data, we use the 0.5° × 0.5° temperature and precipitation data from Méndez and Magaña (2010), who use a Cressman-type interpolation scheme and incorporate meteorological data from the USA, Mexico, and Central America. We also calculate the Standardized Precipitation Index (SPI; Guttman, 1999) and Palmer Drought Severity Index (PDSI; Alley, 1984; Palmer, 1965). The SPI reflects the probability of receiving a given amount of precipitation over a given period of time, while PDSI incorporates the influence of not only precipitation but also evapotranspiration and storage into a widely used index of available moisture (c.f. Cook et al., 2004, 2010b; Touchan et al., 2010). We use the gridded sea surface temperature anomalies (SSTA) from Kaplan et al. (1998) in order to evaluate the potential influence of broad-scale forcing. We additionally compare our chronology with several existing Mexican chronologies from the International Tree-Ring Database (ITRDB). Our assessment of both local and regional climate response is based on the Pearson product-moment correlation.

Results and discussion

A. guatemalensis forms annual rings with a xylem morphology typical of conifers (Figure 3). Our pilot chronology consists of 56 cores from 27 trees in three individual stands and spans the time period 1699 to 2010 ce (Figure 4a). The mean interseries correlation is 0.523 and mean sensitivity is 0.244 (Fritts, 1976; Holmes, 1983), indicating successful crossdating and suggesting a common climate signal. The mean segment length for this set of cores is 125 years. The expressed population signal (Cook and Kairiūkstis, 1990; Wigley et al., 1984) for the master chronology is greater than 0.85 back to 1850 and stable at approximately 0.80 through to the late 18th century, while prior to that the sample depth is currently too small to be representative of the large-scale climate signal (Figure 4b–d). Encouragingly, and despite the distance between sites and difference in elevation, we do not detect any spatial dependence in the strength of the individual between-tree correlations. MSJ samples crossdate successfully with the higher elevation BDR and PAC sites. In those few instances where individual cores could not be successfully crossdated (n = 4), the cause was either a highly compressed section in the outer portion of the tree or complacent growth in juvenile trees.

Figure 3.

Ring morphology in Abies guatemalensis. Upper panel (a) shows a typical series of rings. In the lower panel (b), the arrow indicates a micro-ring. Also evident to the left in the lower photo is earlywood cellular damage with traumatic resin ducts from an unknown cause.

Figure 4.

Cuchumatanes Abies guatemalensis chronology. (a) Complete residual (prewhitened) master chronology, (b) the expressed population signal (EPS) and (c) running interseries correlation (r) in 32 year moving windows overlapped 16 years. (d) Sample size. (b)–(d) show the statistic calculated both for all radii and for trees. (e) Age-aligned raw ring-width series show a negative exponential growth curve. (f) The raw ring-width data.

Our mean interseries correlation (0.523) indicates robust crossdating and common signal amongst trees, and is similar to that reported for Taxodium mucronatum by Stahle et al. (2011b) (0.48) and for Pinus hartwegii at Nevado de Colima, Mexico by Biondi (2001) (0.546), and by Huante et al. (1991) for A. religiousa in Michoacan, Mexico (0.55). Crossdating revealed a relatively small percentage of locally absent rings, about 0.06%. Micro-rings, however, were common (see Figure 3) and provided for several good marker year sequences. Exceptionally wide rings are also present and are useful for crossdating. False rings are also quite common, in many cases requiring careful within-tree comparison to confidently identify. In some years in some trees, we observed cellular damage or trauma (Figure 3); however, while these provided dating markers within the radii of individual trees, we have not yet observed any large-scale coherence within a stand or across sites to suggest a large-scale climate cause for these features. Instead, these may represent trauma or damage only to individual trees, perhaps from geomorphic or meteorological disturbances (e.g. Stoffel, 2008).

Comparisons with the limited – and potentially inhomogeneous – local climate data (Figure 5) suggest that one of the strongest controls on ring width variability in A. guatemalensis in the Cuchumatanes is February precipitation (r = 0.43, p = 0.02). This also appears consistent with the observation that February monthly rainfall has the highest coefficient of variation in the Todos Santos record (Figure 2). When considered on a seasonal basis, total January and February rainfall is also significantly correlated (r = 0.45, p = 0.01) with tree growth. Correlations with temperature are consistently negative, although individually only July temperature meets the significance threshold (r = −0.55, p = 0.01). The broad patterns of negative correlations also results in statistically significant negative correlations with June–July (r = −0.51, p = 0.02), January through June (half year, r = −0.54, p = 0.01), and March though October (‘warm’ or growing season, r = −0.53, p = 0.01) mean temperatures. There are positive relationships between the three-month SPI and ring width, with significant correlations in March (r = 0.41, p = 0.03) and April (r = 0.38, p = 0.04). Boreal spring (March, April, and May, r = 0.52, p = 0.02) and March through October PDSI (warm season, r = 0.46, p = 0.03) are significantly correlated with the master chronology.

Figure 5.

Pearson correlations between the residual master chronology and local climate data from Todos Santos Cuchumatán (15.51°N, 91.60°W, 2480 m), (a) precipitation, (b) temperature, (c) the Standardized Precipitation Index (SPI), and (d) the Palmer Drought Severity Index (PDSI). The latter two metrics are derived from the local climate variables. Dashed red lines indicate the p<0.05 confidence level. For seasonal data, ‘1/2’ refers to the first half of the year (January through June) and ‘W’ to the period from March to October. No significant correlations are found between tree growth and prior year local climate variables (not shown). Colour figure available online.

Gridded climate data products provide an opportunity to extend the comparison of our tree-ring width chronology over the entire 20th century using a validated and homogeneous data set (Figure 6). We observe significant positive correlations between ring width and late winter through early boreal spring (February through April) precipitation over Guatemala, the Yucatan, and southern and eastern coastal Mexico (r = 0.46). These results are similar to those from Huante et al. (1991), who reported a spring precipitation signal in A. religiosa from Michoacan. Correlations between ring width and spring (March and April) temperature are weakly negative (r = −0.17, p < 0.1), with the strongest field correlations to the south through Guatemala, Honduras, El Salvador, and Nicaragua (r = −0.34, p < 0.01). The significant negative correlation with July temperature observed in the local meteorological data is apparently muted in the longer gridded regional data. Correlations with the PDSI field are positive over the tree-ring sites in Guatemala (r = 0.41, p < 0.01), as well as coastal Honduras. We observe a similar PDSI correlation dipole as Stahle et al. (2011b) between Mesoamerica and the southwestern USA. Collectively, the comparisons with local and regional climate data suggest that soil moisture conditions toward the end of the dry season determine interannual variations in ring-width. We hypothesize that this may reflect the influence of late dry season moisture on the timing of the initiation of annual growth, since cores collected in April or May already show several rows of earlywood in some years.

Figure 6.

Spatial correlation between the residual master chronology and gridded (a) precipitation, (b) temperature, and (c) the PDSI derived from them. The 0.5° gridded climate data are from Méndez and Magaña (2010). The season of the comparison is indicated at the top of each plot.

Our chronology is weakly although significantly correlated (r = 0.19, p = 0.02 n = 142, 1865–2006 ce) with the closest existing chronology in the International Tree-Ring Data Bank, from the Rancho Nuevo Pinus montezumae collected by Acuna-Soto and colleagues (ITRDB #MEXI042). The Cuchumatanes chronology also is weakly although significantly correlated (r = 0.16, p = 0.04, n = 159, 1850–2008 CE) with the new millennium-length Barranca de Amealco Taxodium mucronatum from Queretaro in central Mexico developed by Stahle et al. (2011b), which contains a July PDSI climate signal. These relationships improve if first differenced values are used, indicating similar year-to-year anomalies are driving coherence between the sites. The relationship with Barranca de Amealco is slightly improved if only ENSO warm phase (r = 0.40, p = 0.02, n = 30) or cold phase (r = 0.35, p = 0.03, n = 34) years are considered, suggesting stronger tropical Pacific SST anomalies result in larger and spatially coherent hydroclimate anomalies that are reflected in both chronologies, despite their distance. These years appear therefore to account for the similarity in the chronologies’ relationship to the sea surface temperature field (Figure 7). Our Cuchumatanes chronology is significantly negatively correlated with boreal winter (December through March) eastern Pacific SSTs, indicating wider rings during cold ENSO (La Niña) events. The correlation with the NIÑO3 index itself is −0.44 (p < 0.01), and correlations exceed −0.5 in some regions of the Pacific. However, not all ENSO events are recorded by our chronology. Particularly narrow years are observed simultaneously with or following the peak of some El Niño years (e.g. 1973, 1998, and several of the late 19th century events, c.f. Davis, 2001), but not others (e.g. 1983). While the sign of this relationship is generally consistent with observations based on instrumental observations (e.g. Enfield and Mestas-Nuñez, 2000; Ropelewski and Halpert, 1986) and modeling (Seager et al., 2009), and is quite similar to that observed by Stahle et al. (2011a), the relative instability of the relationship could reflect competing seasonal influences from both Pacific and Atlantic teleconnections (Enfield and Mestas-Nuñez, 2000; Giannini et al., 2000, 2001). There are simultaneous positive signed correlations in the Caribbean Sea, which may indicate multiple sea surface temperature influences on the local climate. The correlation between our A. guatemalensis chronology and Pacific SSTs may additionally reflect the influence of increased land surface temperatures, and therefore enhanced evapotranspiration, on the Pacific slope of Central America during El Niño events (Karmalkar et al., 2011). Brienen et al. (2010) report a significant association between a ring width chronology spanning 1970 to 2007 from Mimosa acantholoba in Oaxaca, Mexico and Atlantic as well as tropical Pacific sea surface temperatures. That chronology, however, has a predominantly wet season (June through October) local precipitation response and showed the strongest correlations with tropical Atlantic SSTs. In general, the lack of long instrumental records from the Cuchumatanes themselves impedes a better understanding of how ENSO influences the climate of the region. Despite significant correlations with tropical Pacific sea surface temperatures, the relationship between ENSO and the growth of A. guatemalensis in the Cuchumatanes does not appear to be a straightforward one.

Figure 7.

Spatial correlation between the residual master chronology and gridded Kaplan (Kaplan et al., 1998) gridded winter (December–March) sea surface temperature anomalies (SSTA).

Conclusions

Abies guatemalensis in the high elevations of the Cuchumatanes of Guatemala forms annual rings that can be successfully crossdated. Interannual variability in ring widths is most strongly influenced by early growing season moisture, as determined by late boreal winter–spring precipitation and spring–summer temperatures. The climate signal can be detected using both limited local observations and as well as longer but regional-scale gridded data. We have shown robust crossdating within and between trees, as well as across several sites over an elevation difference of ~400 m. Collectively, our findings here demonstrate that this species can be used to characterize broadscale hydroclimate variability and used for paleoclimate reconstruction in the region. Tree growth sensitivity to late dry season moisture suggests this species can potentially be used to characterize hydroclimate conditions prior to the onset of the rainy season, a time of particular vulnerability for rural populations and agriculturalists in western Guatemala.

Although we have shown that crossdating is robust and a significant climate signal is present, when developing chronologies for climate reconstructions we recommend substantial replication and high sample size to account for the somewhat lower interseries correlation compared with those seen for temperate region semi-arid conifers. A remaining challenge is to identify, collect, and date older individual trees that can be used to extend climate reconstructions further back in time. Relatively little is known about the ecophysiology of this species, and so additional research is also required to understand the timing of growth initiation and dormancy and the association between monthly climate and intra-annual growth patterns. We anticipate that information about the climate response, growth rates, and age structure of this species will also aid in protection, conservation, and restoration efforts (Andersen et al., 2006; Kollmann et al., 2008).

Footnotes

Acknowledgements

We are grateful to the communities of La Ventosa, Todos Santos, and Puerta al Cielo for their valuable assistance in locating and sampling trees in the Cuchumatanes. In particular, we thank Geronimo Ramirez and his family for their continued help and enthusiasm. We also thank Gary Lavanchy, Eliot Andre, Krissy Scommegna, Jesse Dann and Shannon Sullivan for additional field assistance, and Dario Martin-Benito, Caroline Leland, and Laia Andreu-Hayles for additional laboratory assistance and advice. We thank the contributors of the International Tree-Ring Data Bank, IGBP PAGES/World Data Center for Paleoclimatology, NOAA/NCDC Paleoclimatology Program, Boulder, Colorado, USA, for making their data available. We are very grateful to Pauline Décamps and Fernando Mejía of the Unicornio Azul for their generous hospitality and support.

Funding

This research is funded by a grant from the United States National Science Foundation (NSF) Geography and Spatial Science Program (BCS 0852652). Diego Pons was further supported by a LASPAU Climate Change Fellowship from the United States Department of State. Diego Pons and Edwin Castellanos’ collaboration with LDEO and the development of the Universidad del Valle de Guatemala Tree Ring Laboratory were also supported by a Cross Cutting Initiatives grant from Columbia University’s Earth Institute. Christopher Chopp was supported by Columbia University’s Summer Research Program for Science Teachers. This is LDEO Contribution 7597.

References

Acuna-Soto

Stahle

Cleaveland

(2002) Megadrought and megadeath in 16th century Mexico. Emerging Infectious Diseases 8: 360–362.

Aguirre

de Pöll

(2007) Trees in the Life of the Maya World. Fort Worth TX: Botanical Research Institute of Texas.

Alley

(1984) The Palmer Drought Severity Index: Limitations and assumptions. Journal of Applied Meteorology 23: 1100–1109.

Anchukaitis

Evans

(2010) Tropical cloud forest climate variability and the demise of the Monteverde Golden Toad. Proceedings, US National Academy of Science 107(11): 5036–5040, doi: 10.1073/pnas.0908572107.

Anchukaitis

Evans

Wheelwright

. (2008) Stable isotope chronology and climate signal in neotropical montane cloud forest trees. Journal of Geophysical Research 113(G03030), doi: 10.1029/2007JG000613.

Andersen

Sørensen

Cordova

JPP

. (2006) Conservation and utilisation of Abies guatemalensis Rehder (Pinaceae) – An endangered endemic conifer in Central America. Biodiversity and Conservation 15: 3131–3151.

Ault

St George

(2010) The magnitude of decadal and multidecadal variability in North American precipitation. Journal of Climate 23(4): 842–850, doi:10.1175/2009JCLI3013.1.

Barnett

Pierce

(2009) Sustainable water deliveries from the Colorado River in a changing climate. Proceedings, National Academy of Science USA 106(18): 7334.

Bell

Cook

Anchukaitis

. (2011) Repurposing climate reconstructions for drought prediction in Southeast Asia. Climate Change 106: 691–698.

10.

Biondi

(2001) A 400-year tree-ring chronology from the tropical treeline of North America. Ambio 30: 162–166.

11.

Biondi

Hartsough

Galindo Estrada

(2005) Daily weather and tree growth at the tropical treeline of North America. Arctic, Antarctic, and Alpine Research 37(1): 16–24.

12.

Bradley

Keimig

Diaz

(2004) Projected temperature changes along the American cordillera and the planned GCOS network. Geophysical Research Letters 31(16): 10.1029/2004GL020229.

13.

Brienen

RJW

Lebrija-Trejos

Zuidema

. (2010) Climate–growth analysis for a Mexican dry forest tree shows strong impact of sea surface temperatures and predicts future growth declines. Global Change Biology 16(7): 2001–2012.

14.

Buckley

Anchukaitis

Penny

. (2010) Climate as a contributing factor in the demise of Angkor, Cambodia. Proceedings of the National Academy of Sciences 107(15): 6748–6752.

15.

Buckley

Duangsathaporn

Palakit

. (2007) Analyses of growth rings of Pinus merkusii from Lao P.D.R. Forest Ecology and Management 253: 120–127.

16.

Cleaveland

Stahle

Therrell

. (2003) Tree-ring reconstructed winter precipitation and tropical teleconnections in Durango, Mexico. Climate Change 59: 369–388.

17.

Cook

Kairiūkstis

(eds) (1990) Methods of Dendrochronology: Applications in the Environmental Sciences. Doredrecht: Kluwer Academic.

18.

Cook

(1985) A time series approach to tree-ring standardization. PhD Thesis, University of Arizona, Tucson.

19.

Cook

Peters

(1981) The smoothing spline: A new approach to standardizing forest interior tree-ring width series for dendroclimatic studies. Tree-Ring Bulletin 41: 45–53.

20.

Cook

Peters

(1997) Calculating unbiased tree-ring indices for the study of climatic and environmental change. The Holocene 7: 361–370.

21.

Cook

Anchukaitis

Buckley

. (2010b) Asian monsoon failure and megadrought during the last millennium. Science 328(5977): 486–489, 10.1126/science.1185188.

22.

Cook

Meko

Stahle

. (1999) Drought reconstructions for the continental United States. Journal of Climate 12(4): 1145–1162.

23.

Cook

Seager

Heim

. (2010a) Megadroughts in North America: Placing IPCC projections of hydroclimatic change in a long-term paleoclimate context. Journal of Quaternary Science 25(1): 48–61.

24.

Cook

Woodhouse

Eakin

. (2004) Long-term aridity changes in the western United States. Science 306(5698): 1015–1018.

25.

Davis

(2001) Late Victorian Holocausts: El Niño Famines and the Making of the Third World. New York: Verso.

26.

Diaz

Therrell

Stahle

. (2002) Chihuahua (Mexico) winter–spring precipitation reconstructed from tree-rings, 1647–1992. Climate Research 22: 237–244.

27.

Enfield

Mestas-Nuñez

(2000) Global modes of ENSO and non-ENSO sea surface temperature variability and their associations with climate. In: Diaz

Markgraf

(eds) El Niño and the Southern Oscillation: Multiscale Variability and Global and Regional Impacts. Cambridge: Cambridge University Press, pp. 89–112.

28.

Evans

Schrag

(2004) A stable isotope-based approach to tropical dendroclimatology. Geochimica Cosmochimica Acta 68(16): 3295–3305, 10.1016/j.gca.2004.01.006.

29.

February

Stock

(1998) An assessment of the dendrochronological potential of two Podocarpus species. The Holocene 8(6): 747–750.

30.

Fritts

(1976) Tree Rings and Climate. New York: Academic Press.

31.

Giannini

Kushnir

Cane

(2000) Interannual variability of Caribbean rainfall, ENSO, and the Atlantic Ocean. Journal of Climate 13(2): 297–311.

32.

Giannini

Kushnir

Cane

(2001) Seasonality in the impact of ENSO and the North Atlantic high on Caribbean rainfall. Physics and Chemistry of the Earth Part B – Hydrology Oceans and Atmosphere 26(2): 143–147.

33.

Guttman

(1999) Accepting the standardized precipitation index: A calculation algorithm. Journal of the American Water Resources Association 35(2): 311–322.

34.

Holmes

(1983) Computer assisted quality control in tree-ring dating and measurement. Tree-Ring Bulletin 44: 69–75.

35.

Huante

Rincon

Swetnam

(1991) Dendrochronology of Abies religiosa in Michoacan, Mexico. Tree-Ring Bulletin 51: 15–28.

36.

Islebe

Velazquez

Cleef

(1995) High elevation coniferous vegetation of Guatemala. A phytosociological approach. Vegetatio 116(1): 7–23.

37.

Kaplan

Cane

Kushnir

. (1998) Analyses of global sea surface temperature 1856–1991. Journal of Geophysical Research 103(C9): 18,567–18,589.

38.

Karmalkar

Bradley

Diaz

(2011) Climate change in Central America and Mexico: Regional climate model validation and climate change projections. Climate Dynamics 37(3–4): 605–629, doi:10.1007/s00382-011-1099-9.

39.

Kollmann

Córdova

JPP

Andersen

(2008) Factors limiting regeneration of an endangered conifer in the highlands of Guatemala. Journal for Nature Conservation 16: 146–156.

40.

Magaña

Amador

Medina

(1999) The midsummer drought over Mexico and Central America. Journal of Climate 12(6): 1577–1588.

41.

Méndez

Magaña

(2010) Regional aspects of prolonged meteorological droughts over Mexico and Central America. Journal of Climate 23(5): 1175–1188.

42.

Milly

PCD

Betancourt

Falkenmark

. (2008) Stationarity is dead: Whither water wanagement? Science 319(5863): 573.

43.

Palmer

(1965) Meteorological Drought. Technical Report, U.S. Weather Bureau Research Paper 45.

44.

Pederson

Fagre

Gray

. (2004) Decadal-scale climate drivers for glacial dynamics in Glacier National Park, Montana, USA. Geophysical Research Letters 31(12): L12,203.

45.

Pohl

Therrell

Blay

. (2003) A cool-season precipitation reconstruction for Saltillo, Mexico. Tree-Ring Research 59(1): 11–19.

46.

Ropelewski

Halpert

(1986) North American precipitation and temperature patterns associated with the El Niño/Southern Oscillation (ENSO). Monthly Weather Review 114: 2352–2362.

47.

Roy

Lachniet

(2010) Late Quaternary glaciation and equilibrium-line altitudes of the Mayan Ice Cap, Guatemala, Central America. Quaternary Research 74(1): 1–7, 10.1016/j.yqres.2010.04.010.

48.

Schulman

(1944) Dendrochronology in Mexico. Tree-Ring Bulletin 10(3): 18–24.

49.

Scott

(1966) Dendrochronology in Mexico. Papers of the Laboratory of Tree-Ring Research, The University of Arizona Press.

50.

Seager

Ting

Davis

. (2009) Mexican drought: An observational modeling and tree ring study of variability and climate change. Atmósfera 22(1): 1–31.

51.

Speer

Orvis

Grissino-Mayer

. (2004) Assessing the dendrochronological potential of Pinus occidentalis Swartz in the Cordillera Central of the Dominican Republic. The Holocene 14(4): 563–569.

52.

Stahle

Burnette

Diaz

. (2011a) Pacific and Atlantic influences on Mesoamerican climate over the past millennium. Climate Dynamics 10.1007/s00382–011–1205-z.

53.

Stahle

D’Arrigo

Krusic

. (1998a) Experimental dendroclimatic reconstruction of the Southern Oscillation. Bulletin of the American Meteorological Society 79(10): 2137–2152.

54.

Stahle

Diaz

Burnette

. (2011b) Major mesoamerican droughts of the past millennium. Geophysical Research Letters 38(5): L05,703.

55.

Stahle

Villanueva-Diaz

Cleaveland

. (2000a) Recent tree-ring research in Mexico. In: Roig

(ed.) Dendrocronologia en America Latina. Mendoza: Editorial Universidad Nacional de Cuyo, pp. 285–305.

56.

Stahle

(1999) Useful strategies for the development of tropical tree-ring chronologies. IAWA Journal 20: 249–253.

57.

Stahle

Cleaveland

Blanton

. (1998b) The lost colony and jamestown droughts. Science 280: 564–567.

58.

Stahle

Cook

Cleaveland

. (2000b) Tree-ring data document 16th century megadrought over North America. EOS Trans AGU 81(12): 121.

59.

Stahle

Fye

Cook

. (2007) Tree-ring reconstructed megadroughts over North America since A.D. 1300. Climatic Change 83(1–2): 133–149.

60.

Stoffel

(2008) Dating past geomorphic processes with tangential rows of traumatic resin ducts. Dendrochronologia 26(1): 53–60.

61.

Stokes

Smiley

(1968) An Introduction to Tree-Ring Dating. Chicago IL: University of Chicago Press.

62.

Szejner

(2011) Tropical dendrochronology: Exploring tree-rings of Pinus oocarpa in eastern Guatemala. Master’s Thesis, Faculty of Forestry and Forest Ecology Sciences, Georg-August-Universität Göttingen.

63.

Therrell

Stahle

Cleaveland

. (2002) Warm season tree growth and precipitation over Mexico. Journal of Geophysical Research 107(4205): 10.1029/2001JD000851.

64.

Therrell

Stahle

Diaz

. (2006) Tree-ring reconstructed maize yield in central Mexico: 1474–2001. Climatic Change 74(4): 493–504.

65.

Therrell

Stahle

Soto

(2004) Aztec drought and the ‘Curse of One Rabbit’. Bulletin of the American Meteorological Society 85(9): 1263–1272.

66.

Touchan

Anchukaitis

Meko

. (2008) Long term context for recent drought in northwestern Africa. Geophysical Research Letters 35(13): L13,705.

67.

Touchan

Anchukaitis

Meko

. (2010) Spatiotemporal drought variability in northwestern Africa over the last nine centuries. Climate Dynamics doi: 10.1007/s00382–010–0804–4.

68.

Veblen

(1978) Forest preservation in the Western Highlands of Guatemala. Geographical Review 68(4): 417–434.

69.

Villanueva-Diaz

Stahle

Luckman

. (2007) Winter–spring precipitation reconstructions from tree rings for northeast Mexico. Climate Change 83(1): 117–131.

70.

Vorosmarty

Green

Salisbury

. (2000) Global water resources: Vulnerability from climate change and population growth. Science 289(5477): 284–288. doi: 10.1126/science.289.5477.284.

71.

Wigley

TML

Briffa

Jones

(1984) On the average value of correlated time series, with applications in dendroclimatology and hydrometeorology. Journal of Climate and Applied Meteorology 23: 201–213.

72.

Woodhouse

Lukas

(2006) Multi-century tree-ring reconstructions of Colorado streamflow for water resource planning. Climatic Change 78(2): 293–315.

73.

Woodhouse

Gray

Meko

(2006) Updated streamflow reconstructions for the Upper Colorado River basin. Water Resou rces Research 42(W05415) doi:10.1029/2005WR004455.

74.

Yamaguchi

(1991) A simple method for cross-dating increment cores from living trees. Canadian Journal of Forest Research 21: 414–416.