High altitude C4 grasslands in the northern Andes: relicts from glacial conditions?
Introduction
Reconstructions of climate change during the Quaternary of tropical South America are mainly based on vegetation change documented in pollen records. Records from tropical mountains are of particular interest for paleoclimate reconstructions as vegetation belts migrate along altitudinal gradients during glacial–interglacial cycles. Stable carbon isotopic analyses (δ13C) of sedimentary organic matter have been applied to cores from tropical mountain ecosystems. These studies have provided a new proxy of environmental change, which gives a special role to C4 plants and partial atmospheric CO2 pressure (pCO2) (Street-Perrott, 1994, Street-Perrott et al., 1997, Ficken et al., 1998, Boom and Marchant, 2001). Earlier papers dealing with δ13C of tropical records use aridity to explain expansion of C4 grasslands (Aucour and Hillaire-Marcel, 1994, Giresse et al., 1994, Quade et al., 1989, Sukumar et al., 1993, Talbot and Johannesen, 1992).
Among vascular plants, three different major photosynthetic routes can be distinguished: C3, C4 and CAM photosynthesis. The most common among present-day plants is the C3 pathway. The C4 pathway enables plants a higher water use efficiency (WUE) and a more effective CO2 uptake, because they use a CO2 concentrating mechanism (Leegood, 1999). At present, many C4 plants can be found among the monocot families Cyperaceae (sedges) and Poaceae (grasses). According to Watson and Dallwitz (1992), 407 out of 799 Poaceae genera contain C4 species, most of them are characteristic elements of tropical to temperate areas with abundant warm-season precipitation. WUE of CAM plants is even better than that of C4 plants, and thus many of them can be found in hot and arid environments, where other plants cannot grow. Some CAM plants are important components of the high Andean vegetation, such as Bromeliaceae (e.g. Puya sp.). However, CAM photosynthesis can also be found in aquatic plants and epiphytes. The occurrence of CAM in Isoetes, an important plant from Andean lakes, may even point to a very remote origin in the evolution of CAM (Keeley, 1998), while the C4 pathway is believed to have evolved during the mid-Cretaceous (C4 grasses: Brown and Smith, 1992; C4 plants: Bocherens et al., 1993). However, little is known about their behavior in response to Quaternary climate change, neither do they make a significant contribution to the pollen based vegetation reconstructions for tropical South America.
Most of the modern C4 Poaceae are found at places where, during the growing season, average monthly temperature exceed 22°C and rainfall is more than 25 mm/month (Ehleringer, 1997, Collatz et al., 1998). By using established models for C3 and C4 photosynthesis, a relationship that describes the competitive balance between the two was proposed (Ehleringer, 1997, Collatz et al., 1998). The authors defined the point where both types of photosynthesis are equally effective in assimilating carbon, in terms of a crossover point (or crossover temperature), which is a function of pCO2. Whenever one type is more effective than the other, that type wins the competition (Fig. 1). Thus under low pCO2, C4 grasses will be able to dominate over the C3 species; even in colder areas where, under current pCO2 conditions, C4 grasses are only of minor importance compared to C3 species. Lowered atmospheric pCO2 levels in the Miocene are believed to have resulted in the appearance of C4 dominated grasslands (Cerling et al., 1993, Cerling et al., 1997, Ehleringer, 1997, Latorre et al., 1997, Quade and Cerling, 1995, Quade et al., 1995).
Independent from the previous model, the C4 plants will also have an advantage over the C3 plants under low atmospheric CO2 concentrations. To maintain a constant influx of CO2, plants will produce more stomata per surface unit (Beerling and Chaloner, 1992, Beerling and Chaloner, 1994, Beerling et al., 1995, Wagner, 1998, Wagner et al., 1999). This will cause increased water loss through evaporation and resulting in water stress. Altered atmospheric pCO2 even influences pure C3 plant communities. Jolly and Haxeltine (1997) used a BIOME-3 modeling experiment (Haxeltine and Prentice, 1996) at the site of Kashiru bog in tropical East Africa to show that the montane forest at 2240 m elevation can be completely replaced by ericaceous scrub vegetation by lowering only the pCO2 to last glacial maximum (LGM) values while maintaining a constant temperature.
Gas analysis of air bubbles from the Vostok ice core, covering the last 420,000 years, shows that CO2 levels have varied significantly during the late Quaternary (Barnola et al., 1987, Petit et al., 1999), showing a significant drop of atmospheric CO2 concentrations from 280 ppmV during interglacial time (modern pre-industrial value) to 180 ppmV during the LGM. This pattern is consistent with at least three older glaciations recorded in the Vostok ice core.
Until recently, lower temperatures during the glaciations have been taken as the main cause for the lower altitudinal position of the upper forest line in the northern Andes (e.g. Van der Hammen and González, 1960; Hooghiemstra, 1984; Van der Hammen and Cleef, 1986). The objective of this paper is to identify the impact of a significant drop in the concentration of atmospheric carbondioxide on the altitudinal vegetation distribution in the Colombian Andes, the C4–C3 grass crossover in particular. With stable carbon isotopic analysis, a Colombian high-altitude record is investigated on the presence of C4 plants in the past. We aim to provide a sketch of potential plant communities that could have occurred under glacial low atmospheric pCO2 and which do not occur under Holocene conditions and finally we suggest the presence of a possible relict of these vegetation types.
Section snippets
Location and current altitudinal vegetation zonation
We focus on the northern Andes, more specifically the area around the high plain of Bogotá, which is situated at 2550 m elevation in the Eastern Cordillera, Colombia. The basin, which represents the bottom of a dry paleo-lake, accumulated an extraordinary sequence of lake sediments during the Quaternary. Core Funza-II was drilled at the site where the basin was expected to have maximum depth. From the tropical lowlands to the temperate high altitudes, six distinct modern vegetation belts are
Methods
Aliquots of about 0.5 g of sediment from Funza-II were decalcified for isotopic determinations of organic matter by overnight reaction with 5N HCl at room temperature. The acidic solution was removed by centrifugation and aspiration. The insoluble residue was neutralized using repeated suspension in distilled/de-ionized water followed by centrifugation and vacuum aspiration. Neutralized residues were dried overnight at 60°C, mixed with about 1.5 g of CuO, and put into 9 mm quartz tubes. The tubes
Stable carbon isotopic record of Funza-II
The stable isotopic record of the total organic matter (TOC) is given in Fig. 3. Since the Funza-II sediments are lacustrine, TOC include both terrestrial and lacustrine components. A detailed analysis of the Funza-II δ13CTOC (Mora et al., in press) demonstrates that the δ13CTOC shows mainly changes in terrestrial C3 and C4 dominated ecosystems (Fig. 3). Many of the glacial periods show enrichments in δ13CTOC. C4 plants have characteristic average δ13CTOC values of −12‰ (Bender, 1971) as
Conclusions
Based on theoretical considerations and stable carbon isotope data, we propose that C4 grasslands must have developed during (parts of) glacial periods in the basin of Bogotá. There is sufficient information available to accept that replacement of C3 grasslands by C4 grasslands is climatically controlled and thus has implications for climatic reconstruction of the area. Changes in temperature, precipitation, and pCO2 play an important role in the altitudinal zonation of vegetation belts.
Acknowledgements
The authors thank M. Bush, M. Cabido and H. Visscher for useful comments. We thank J.J. Boon (FOM/ Hugo de Vries Laboratory, Amsterdam), R. Marchant (Hugo de Vries Laboratory, Amsterdam) and J.S. Sinninghe Damsté (NIOZ, Texel) for constructive comments on the manuscript, S. Schouten (NIOZ, Texel) for advise and support, and J.W. de Leeuw (NIOZ, Texel) for making facilities at the NIOZ Institute available. J.C. Berrio (Bogotá/Amsterdam), the family Berrio in Puntalarga (Boyacá), and C. Rodriguez
References (63)
- et al.
Late Quaternary biomass changes from 13C measurements in a highland peatbog from Equatorial Africa (Burundi)
Quat. Res.
(1994) - et al.
Atmospheric CO2 changes since the last glacial maximum: evidence from the stomatal density record of fossil leaves
Rev. Palaeobot. Palynol.
(1994) Variations on the 13C ratios of plants in relation to the pathway of photosynthetic carbon dioxide fixation
Phytochemistry
(1971)- et al.
Late Quarternary palaeoenvironments in Lake Barombi Mbo (West Cameroon) deduced from pollen and carbon isotopes of organic matter
Palaeogeogr. Palaeoclimatol. Palaeoecol.
(1994) - et al.
The 14C dates and 13C/12C ratios of soil organic matter as a means of studying the past vegetation in intertropical regions: examples from Colombia (South America)
Palaeogeogr. Palaeoclimatol. Palaeoecol.
(1988) - et al.
Late and middle Pleistocene climatic change and forest development in Colombia: pollen record Funza-II (2–158 m core interval)
Palaeogeogr., Palaeoclimatol., Palaeoecol.
(1994) - et al.
Late Quaternary vegetation history and paleoecology of Laguna Pedro Palo (subandean forest belt, Eastern Cordillera. Colombia)
Rev. Palaeobot. Palynol.
(1993) - et al.
The expansion of C4 grasses and global change in the late Miocene: Stable isotope evidence from the Americas
Earth Planet. Sci. Lett.
(1997) Carbon isotope fractionation in plants
Phytochemistry
(1981)- et al.
Expansion of C4 grasses in the late Miocene of Northern Pakistan: evidence from stable carbon isotopes in palaeosols
Palaeogeogr. Palaeoclimatol. Palaeoecol.
(1995)