Amazonian mangrove dynamics during the last millennium: The relative sea-level and the Little Ice Age
Introduction
There is evidence that the sea-level has increased since the end of the Little Ice Age (Ekman, 1999). The resulting global mean sea-level rise of about 18 cm during the last century (Gornitz, 1995) has produced displacements of coastal ecosystems (Chappell, 1990, IPCC (Intergovernmental Panel on Climate Change), 1996, Crooks and Turner, 1999). As a consequence, estuaries have migrated landwards at rates of approximately 10 m/yr, whereas open-coast landforms may exhibit long-shore migration rates of 50 m/yr (Pethick, 2001). On the Pacific margin of Canada, the relative sea-level rise of up to 40 cm has resulted in as much as 12 m of coastal retreat (Barrie and Conway, 2002). In Brazil, tidal records obtained over the last 50 years show a general rise in relative sea-level (Pirazolli, 1986, Mesquita and Harari, 1983, Mesquita and Leite, 1985, Silva and Neves, 1991, Silva, 1992, Aubrey et al., 1988).
The effects of this sea-level increase become apparent by the net loss of mangroves resulting from coastal erosion in the tropical regions of Senegal, Benin, Côte d'Ivoire, Colombia, Venezuela, and Brazil (Blasco et al., 1996). On the Brazilian coastline effects seem to be reflected particularly by shore erosion (Muehe and Neves, 1995). Southeast of the Amazon River mouth, a significant part of the Pará State coastline has undergone a net loss of mangrove coverage area during the last 25 years. On the other hand, mangroves have invaded herbaceous flats at higher elevations during same period (Cohen and Lara, 2003).
Increasing greenhouse gas concentrations are thought to be the dominant forcing factor of climate change over the last decades (IPCC, 1996), driving temperatures to unprecedented levels (Bradley, 2000), and resulting in varying predictions of rising sea-level rates of + 15 cm by the year 2050 and of + 35 cm by 2100 (Titus and Narayanan, 1995). More recent predictions indicate a sea-level rise between 60 and 100 cm by the end of this century (Douglas et al., 2000).
Also pre-industrial surface temperatures varied significantly, probably provoked by fluctuations of solar activity levels, which produced relatively cold periods during the Little Ice Age (LIA) (Lean and Rind, 1999). The conventional view of the climate history of this period has been traditionally based on European weather records, delineating two distinct time periods, the LIA and the preceding Medieval Warm Period (Fig. 1).
During the LIA, glaciers extended on the Northern Hemisphere (Grove, 2001), e.g., in Switzerland (Röthlisberger et al., 1980), Alaska (Calkin et al., 2001) and in the Canadian Rockies (Luckman, 2000). Based on glacial advances, the LIA occurred during the last six or seven centuries, and ended somewhere between 1850 and 1890 AD (Bradley and Jones, 1992). This cold period was recognized in South America as well by several authors (e.g., Heusser, 1961, Politis, 1984, Hurtado et al., 1985), as causing glacier advances in the Andes (Iriondo and Kröhling, 1995, Malagnino and Strelin, 1996) and aridity in the lowlands of Argentina (Cioccale, 1999) and Venezuela (Iriondo, 1999). In Argentina, this dry period provoked a recession of the fluvial systems (Furlong Cardiff, 1937). Nevertheless, it is not clear whether the conception of the “Little Ice Age” derived from the Northern Hemisphere is applicable to the Southern (Grove, 2001).
The LIA effects have not been explicitly described for Brazilian ecosystems. For the Brazilian southern region, Behling et al. (2004b) have described changes of vegetation patterns as coinciding with the LIA. In northern Brazil, the LIA effects might be recorded on the coastal zones, since over 80% of sediments of the Amazon River discharge is derived from the Andes (Gibbs, 1977). Changes are reflected in the Amazon shelf sedimentation, which registered depositional conditions from at least 200 to 1310 AD and an erosional phase from 1310 to 1860 AD (Sommerfield et al., 1995).
The sediment transport of the Amazon River has produced the longest mud coastline in the world (Kjerfve and Lacerda, 1993). This coastline is colonized by mangroves, which are considered highly susceptible to sea-level fluctuations and climatic changes (Gornitz, 1991). Recently, well established sea-level histories based on coastal sequences have become important in the study and prediction of global climatic, oceanographic and tectonic fluctuations (Pirazolli, 1988, Shennan, 1989).
Land subsidence may produces a local to regional relative sea-level rise (Emery and Aubrey, 1991, Church and Coe, 2003, Mörner, 1999), and, probably, has influenced the development of a jagged coast with numerous bays and estuaries in the Pará coastal zone (Souza Filho, 2000), which can also have been contributing to the mangrove establishment on the Bragança Peninsula.
The purpose of this paper is to study the environmental history of the Bragança Peninsula during the last 1000 years, focusing on the vegetation development in the central part of the peninsula, where boundaries of mangrove and salt marshes occur, and sensitive vegetation changes related to relative sea-level changes can be expected. Thus, propose the LIA-effects in Southern Hemisphere on the basis of changing mangrove distribution. The approach used involves stratigraphy, pollen analysis and radiocarbon dating.
Section snippets
Study area
The study site is located on the Bragança Peninsula, in the eastern Amazon region, between the Maiaú and Caeté bays (00° 46′ 00″– 1° 00′ 00″ S and 46° 36′ 00″–46° 44′ 00″ W, Fig. 2). Modern vegetation of the Bragança Peninsula is represented by the following units: Amazon coastal forest, elevated herbaceous flats (EHF), mangroves, degraded mangroves and restinga (shrub and herb vegetation that occurs on sand plains and on dunes close to the shore line). These units are distributed within
Sampling and sample processing
Nine sediment cores of one meter depth were collected (Fig. 2) using vibracore equipment. The geographical positions of the sediment cores was determined by GPS, and the topographic data were taken by hydrographic devices as described in Cohen et al. (2001). Sediment color was classified according to the Rock-Color Chart (1984).
The cores of the southern (M1, M2, M3 and M4) and northern part (M5, M6, M7 and M8) of the studied transect of 3.2 km length (Fig. 2) were sampled from areas colonized
Stratigraphy
The cores from the northern and southern part of the studied transect (Fig. 2) show significant stratigraphic differences (Fig. 4). Most of the sediment cores have common layers (C1 and C2) only at their base. The core base from the northern and southern part contains gray (N8) fine sand, well sorted, with sand ripples inserted with mm-and cm-thick gray (N5) mud lens, composing the wavy lamination of the C1 layer. This grain size alternation can be related to tidal action (Reineck and
Relationship of tectonic control and relative sea-level
Tectonic movements can produce considerable subsidence or uplift of the coastal zone in a local to regional scale that generate, jointly with other variable, relative sea-level changes (Emery and Aubrey, 1991, Church and Coe, 2003, Mörner, 1999). Marine transgression is decreased to outbalanced by increasing rates of uplift, whilst regressions are enlarged. In a subsiding area, the regressions are decreased to outbalanced by increasing rates of subsidence, whilst the transgressions are
Conclusions
The integration of pollen and stratigraphic analyses, including AMS-radiocarbon dating, was used to identify the inundation regime characteristics of the Bragança Peninsula in northern Brazil during the last millennium. This study suggests two periods of low inundation frequency (P1 and P2), which probably occurred between 1130 and 1510 AD; and 1560 AD and the end of the 19th century, respectively. Likely, this alternation of dry and wet sediment conditions is coupled with relative sea-level
Acknowledgements
We thank the members of the Center for Tropical Marine Ecology in Germany and Núcleo de Estudos Costeiros in Brazil for their support. This study was carried out as a part of the Brazilian-German Cooperation Project Mangrove Dynamics and Management (MADAM), and was financed by the Brazilian National Research Council (CNPq) and the German Ministry for Education and Research (BMBF) under the code 03F0154A. The first author thanks the Deutscher Akademischer Austauschdienst (DAAD) for the
References (100)
- et al.
High resolution seismic stratigraphy and its sedimentological interpretation on the Amazon continental shelf
Continental Shelf Research
(1986) - et al.
Changing coastal levels of South America and the Caribbean region from tide-gauge records
Tectonophysics
(1988) - et al.
Rapid sea-level change and coastal evolution on the Pacific margin of Canada
Sedimentary Geology
(2002) - et al.
Calibration of radiocarbon ages and the interpretation of paleoenvironmental records
Quaternary Research
(1995) - et al.
Studies on Holocene mangrove ecosystem dynamics of the Bragança Peninsula in north-eastern Pará, Brazil
Palaeogeography, Palaeoclimatology, Palaeoecology
(2001) - et al.
Late Quaternary Araucaria forest, grassland (Campos), fire and climate dynamics, studied by high resolution pollen, charcoal and multivariate analysis of the Cambará do Sul core in southern Brazil
Palaeogeography, Palaeoclimatology, Palaeoecology
(2004) - et al.
Spatial and temporal variability in the Holocene sea-level record of the South Australian coastline
Sedimentary Geology
(2002) - et al.
Mangroves as indicators of coastal change
Catena
(1996) Past global changes and their significance for the future
Quaternary Science Reviews
(2000)- et al.
Stratigraphic and micropetrographic occurrences of pyrite in sediments at the confluence of carbonate and peat-forming depositional systems, southern Florida, U.S.A.
Organic Geochemistry
(1995)