New radiocarbon calibration curves, IntCal04 and Marine04, have been constructed and internationally ratified to replace the terrestrial and marine components of IntCal98. The new calibration data sets extend an additional 2000 yr, from 0–26 cal kyr BP (Before Present, 0 cal BP = AD 1950), and provide much higher resolution, greater precision, and more detailed structure than IntCal98. For the Marine04 curve, dendrochronologically-dated tree-ring samples, converted with a box diffusion model to marine mixed-layer ages, cover the period from 0–10.5 cal kyr BP. Beyond 10.5 cal kyr BP, high-resolution marine data become available from foraminifera in varved sediments and U/Th-dated corals. The marine records are corrected with site-specific 14C reservoir age information to provide a single global marine mixed-layer calibration from 10.5–26.0 cal kyr BP. A substantial enhancement relative to IntCal98 is the introduction of a random walk model, which takes into account the uncertainty in both the calendar age and the 14C age to calculate the underlying calibration curve (Buck and Blackwell, this issue). The marine data sets and calibration curve for marine samples from the surface mixed layer (Marine04) are discussed here. The tree-ring data sets, sources of uncertainty, and regional offsets are presented in detail in a companion paper by Reimer et al. (this issue).

A nearshore core (LT03-05) from the north basin of Lake Tanganyika provides diatom, pollen, and sedimentary time series covering the last ca. 3800 yr at 15-36 yr resolution. A chronology supported by 21 AMS dates on terrestrial and lacustrine materials allows us to account for ancient carbon effects on (14)C ages were and to propose refinements of the region's climatic history. Conditions drier than those of today followed after ca. 3.30 ka by an overall wetting trend. Several century-scale climate variations were superimposed upon that trend, with exceptionally rainy conditions occurring 1.70-1.40 ka, 1.15-0.90 ka, and 0.35-0.20 ka. Around 0.55-0.35 ka, during the Sporer Sunspot minimum, drier conditions 0.70-0.55 ka., developed in the northern Tanganyika basin while more humid conditions were registered at Lakes Victoria and Naivasha. This indicates significant variability in the nature and distribution of near-equatorial rainfall anomalies during Much of the Little Ice Age. (C) 2009 University of Washington. Published by Elsevier Inc. All rights reserved

The palynology of two overlapping Holocene cores froth Lake Sibaya in KwaZulu-Natal elucidates the relationship between climate, vegetation and human impact in the region. By means of twenty-one AMS (14)C dates, loss on ignition, and palynological results we established a composite profile. Pollen assemblages include elements of swamp forest (e.g. Rauvolfia, Macaranga), dune forest (e.g. Mimusops), mangrove vegetation (Bruguiera), palmveld (e.g. Phoenix) and bushveld (e.g. Spirostachys. Sclerocarya). Poaceae, aquatics and Cyperaceae are abundant, and fynbos elements like Ericaceae and Restionaceae are rare. Based on comparisons between palynological and archaeological/historical data, the radiocarbon dates seem to show an age error of 50-550 yr, which is probably due to a hardwater reservoir effect. Applying the mean of this error range to our age model suggests that the oldest sediments represent similar to 6750-7100 cal yr BP, that a > 5000 yr hiatus occurs ca. 253 cm depth, and that the upper 253 cm of the composite profile covers the period between similar to 1300-1500 cal yr BP (similar to 450-650 AD, Early Iron Age). and 2004 AD. The Middle Holocene is characterized by high tree pollen values (especially Phoenix) suggesting warm humid conditions. The Early Iron Age is characterized by high Podocarpus percentages that indicate moist but possibly cooler climatic conditions. The upper part of the pollen sequence is characterized by the decrease of Podocarpus. Isoglossa and Celtis and a rise in Spirostachys. Increasing values of cereal pollen and algae might reflect human activity. Zea mays appears similar to 150-300 cal yr BP in the pollen sequence according to the radiocarbon chronology and both archaeological and historical evidence. The curve of Pinus pollen rises to 50-70% at the top of the diagram, reflecting the spread of pine plantations since the 1920's. and Poaceae values decrease. Stoebe and the introduction of neophytes like Ambrosio and Casuarina suggest recent human disturbance. (c) 2008 Elsevier B.V. All rights reserved.

The geochemistry and isotopic composition (H, O, S, O-sulfate, C, Sr) of groundwater from the Nubian Sandstone (Kurnub Group) aquifer in the Negev, Israel, were investigated in an attempt to reconstruct the origin of the water and solutes, evaluate modes of water-rock interactions, and determine mean residence times of the water. The results indicate multiple recharge events into the Nubian sandstone aquifer characterized by distinctive isotope signatures and deuterium excess values. In the northeastern Negev, groundwater was identified with deuterium excess values of similar to 16 parts per thousand, which suggests local recharge via unconfined areas of the aquifer in the Negev anticline systems. The delta O-18(H2O) and delta H-2 values (-6.5 parts per thousand and -35.4 parts per thousand) of this groundwater are higher than those of groundwater in the Sinai Peninsula and southern Arava valley (-7.5 parts per thousand and -48.3 parts per thousand) that likewise have lower deuterium excess values of -10 parts per thousand. Based on the geochemical differences between groundwater in the unconfined and confined zones of the aquifer, a conceptual geochemical model for the evolution of the groundwater in the Nubian sandstone aquifer has been reconstructed. The isotopic composition of shallow groundwater from the unconfined zone indicates that during recharge oxidation of pyrite to SO4 (delta S-34(SO4) similar to-13 parts per thousand; delta O-18(SO4) similar to+7.7 parts per thousand) and dissolution of CaCO3 (Sr-87/Sr-86 similar to 0.70787; delta C-13(DIC) = -3.7 parts per thousand) occur. In the confined zone of the aquifer, bacterial SO4 reduction removes a significant part of dissolved SO42-, thereby modifying its isotopic composition (delta S-34(SO4) similar to-2 parts per thousand; delta O-18(SO4) similar to+8.5 parts per thousand) and liberating dissolved inorganic C that contains little or no radiocarbon (C-14-free) with low delta C-13(DIC) values (<- 12 parts per thousand). In addition to local recharge, the Sr and S isotopic data revealed contribution of external groundwater sources to the Nubian Sandstone aquifer, resulting in further modifications of the groundwater chemical and isotopic signatures. In the northeastern Negev, it is shown that SO4-rich groundwater from the underlying Jurassic aquifer contributes significantly to the salt budget of the Nubian Sandstone aquifer. The unique chemical and isotopic composition of the Jurassic groundwater (delta S-34(SO4) similar to +14 parts per thousand; delta O-18(SO4) similar to 14 parts per thousand; Sr-87/Sr-86 similar to 0.70764) is interpreted as reflecting dissolution of Late Triassic marine gypsum deposits. In the southern Arava Valley the authors postulate that SO4-rich groundwater with distinctively high Br/Cl (3 x 10(-3)) low Sr-87/Sr-86 (0.70734), and high delta S-34(SO4) values (+ 15 parts per thousand) is derived from mixing with underlying brines from the Paleozoic units. The radiocarbon measurements reveal low C-14 activities (0.2-5.8 pmc) in both the northeastern Negev and southern Arava Valley.Taking into account dissolution of carbonate rocks and bacterial SO4 reduction in the unconfined area, estimated mean residence times of groundwater in the confined zone in the northeastern Negev are on the order of 21-38 ka, which sggests recharge predominantly during the last glacial period. The C-14 signal in groundwater from the southern Arava Valley is equally low but due to evidence for mixing with external water sources the residence time estimates are questionable. (c) 2007 Elsevier Ltd. All rights reserved.

PAGES News, Vol.14 • No 3 • December 2006 S ci en ce H ig h lig ht s: 14 C C h ro n ol og y can defi ne (through calculation) what the ‘true’ C age will be (the consensus value) and then we can estimate for each laboratory, whether there is a constant off set (or a bias) from this consensus. The current program of inter-laboratory comparisons has aff orded an opportunity for laboratories to assess their accuracy. In each intercomparison, the consensus values for the unknown age samples was calculated and reported. Figure 2, , shows the off set (and 95% confi dence interval) for individual laboratories based on the dendrochronologically dated samples included in FIRI. The sample dendro-ages were 3200-3239 BC, 3299 3257 BC and 313-294 BC. Conclusions Analyses of results from FIRI and phase 1 of VIRI support the fact that radiocarbon laboratories are generally accurate and precise. The results from FIRI are signifi cant in that they show broad agreement between measurements made in diff erent laboratories on a wide range of materials, and they also demonstrate no statistically signifi cant diff erence between measurements made by radiometric or AMS techniques. As a result of the inter-comparison program, an extensive suite of natural reference materials (such as wood, carbonate, etc) spanning the applied C timescale has been created by the C dating community. These can now be used by C labs to improve their dating accuracy and are thus of great benefi t to the users of C dates.

Multi-proxy high-resolution records relating to climate and dominant vegetation cover have been obtained from a peat deposit retrieved from the Kapsabet swamp in western Kenya. The 4-m long peat sequence provided a continuous sedimentation record spanning ca. 3023 cal yr BP to the present and is representative of the late Holocene changes in the relatively high-altitude part of the Lake Victoria catchment. Paleoenvironmental factors influencing peat formation and organic matter (OM) source inputs in Kapsabet were reconstructed based on total organic carbon, carbon accumulation rate, δ13C, δ15N, C/N ratio, and specific biomarker-based n-alkane ratios. The Kapsabet peat sequence was divided into five stages based on different climatic conditions: Stage 1 (3023–1670 cal yr BP) represents a progression from cool dry to wet conditions. Stage 2 (1670–1187 cal yr BP) is a relatively warm and wet period with increased productivity and high OM input. Stage 3 (1187–625 cal yr BP) represents gradual warming coeval to the late Holocene Medieval Warm Period. Stage 4 (625–188 cal yr BP) is a cool and wet period with high variability in precipitation and hydrological conditions representing the Little Ice Age. Stage 5 (188 cal yr BP to present) represents a relatively cool and wet period that coincides with the expansion of agriculture, particularly in the lowlands. The proxies indicate a progressive change from a forested landscape to an open woodland coeval to a decline in terrigenous inputs and the advent of more wet conditions. The climate on the highlands was less variable than in the lowlands, which underwent several periods of drought and intermittent wet conditions. The changes in the catchment coincided with the expansion of agriculture and land clearance marking increased human activities in the lowlands. Overall, the Kapsabet peat sequence tracks the regional climatic changes in East Africa and marks a promising sedimentary archive for palaeoclimate reconstruction from a region with a paucity of palaeoenvironmental and palaeovegetation histories.

Elemental and lipid analyses were carried out on sediments recovered from Ocean Drilling Program (ODP) Site 942 to provide a 35 kyr record of organic matter input to Amazon Fan sediments. Total organic carbon (TOC) and higher plant biomarker mass accumulation rates were an order of magnitude greater during the last glacial period compared to the current interglacial due to sea-level controlled variations in Amazon River sediment supply. Large maxima were also seen at ∼12 ka, which are most likely due to discharge events. Higher plant n-alkane average chain lengths did not change throughout the record, suggesting consistency in the source vegetation type. The abundance of taraxerol relative to other plant biomarkers increased at ∼12 ka, indicating increased mangrove input due to either higher mangrove productivity or increased erosion of mangrove deposits. The mass accumulation rates (MARs) of some bacterial and eustigmatophyte biomarkers varied closely with those of higher plant biomarkers and so seem to have a non-marine source. Long chain alkenones were present in some of the sediments, generally in very low concentration, indicating dilution of the marine signal with terrestrial organic matter.

Radiocarbon calibration curves are essential for converting radiocarbon dated chronologies to the calendar timescale. Prior to the 1980's numerous differently derived calibration curves based on radiocarbon ages of known age material were in use, resulting in ''apples and oranges'' comparisons between various records (Klein et al., 1982), further complicated by until then unappreciated inter-laboratory variations (International Study Group, 1982). The solution was to produce an internationally-agreed calibration curve based on carefully screened data with updates at 4-6 year intervals (Klein et al., 1982; Stuiver and Reimer, 1986; Stuiver and Reimer, 1993; Stuiver et al., 1998). The IntCal working group has continued this tradition with the active participation of researchers who produced the records that were considered for incorporation into the current, internationally-ratified calibration curves, IntCal04, SHCal04, and Marine04, for Northern Hemisphere terrestrial, Southern Hemisphere terrestrial, and marine samples, respectively (Reimer et al., 2004; Hughen et al., 2004; McCormac et al., 2004). Fairbanks et al. (2005), accompanied by a more technical paper, Chiu et al. (2005), and an introductory comment, Adkins (2005), recently published a ''calibration curve spanning 0-50,000 years''. Fairbanks et al. (2005) and Chiu et al. (2005) have made a significant contribution to the database on which the IntCal04 and Marine04 calibration curves are based. These authors have now taken the further step to derive their own radiocarbon calibration extending to 50,000 cal BP, which they claim is superior to that generated by the IntCal working group. In their papers, these authors are strongly critical of the IntCal calibration efforts for what they claim to be inadequate screening and sample pretreatment methods. While these criticisms may ultimately be helpful in identifying a better set of protocols, we feel that there are also several erroneous and misleading statements made by these authors which require a response by the IntCal working group. Furthermore, we would like to comment on the sample selection criteria, pretreatment methods, and statistical methods utilized by Fairbanks et al. in derivation of their own radiocarbon calibration.

New radiocarbon calibration curves, IntCal04 and Marine04, have been constructed and internationally ratified to replace the terrestrial and marine components of IntCal98. The new calibration data sets extend an additional 2000 yr, from 0–26 cal kyr BP (Before Present, 0 cal BP = AD 1950), and provide much higher resolution, greater precision, and more detailed structure than IntCal98. For the Marine04 curve, dendrochronologically-dated tree-ring samples, converted with a box diffusion model to marine mixed-layer ages, cover the period from 0–10.5 cal kyr BP. Beyond 10.5 cal kyr BP, high-resolution marine data become available from foraminifera in varved sediments and U/Th-dated corals. The marine records are corrected with site-specific 14C reservoir age information to provide a single global marine mixed-layer calibration from 10.5–26.0 cal kyr BP. A substantial enhancement relative to IntCal98 is the introduction of a random walk model, which takes into account the uncertainty in both the calendar age and the 14C age to calculate the underlying calibration curve (Buck and Blackwell, this issue). The marine data sets and calibration curve for marine samples from the surface mixed layer (Marine04) are discussed here. The tree-ring data sets, sources of uncertainty, and regional offsets are presented in detail in a companion paper by Reimer et al. (this issue).

The radiocarbon calibration curve IntCal04 extends back to 26 cal kyr B P. While several high-resolution records exist beyond this limit, these data sets exhibit discrepancies of up to several millennia. As a result, no calibration curve for the time range 26–50 cal kyr BP can be recommended as yet, but in this paper the IntCal04 working group compares the available data sets and offers a discussion of the information that they hold.

A new calibration curve for the conversion of radiocarbon ages to calibrated (cal) ages has been constructed and internationally ratified to replace IntCal98, which extended from 0–24 cal kyr BP (Before Present, 0 cal BP = AD 1950). The new calibration data set for terrestrial samples extends from 0–26 cal kyr BP, but with much higher resolution beyond 11.4 cal kyr BP than IntCal98. Dendrochronologically-dated tree-ring samples cover the period from 0–12.4 cal kyr BP. Beyond the end of the tree rings, data from marine records (corals and foraminifera) are converted to the atmospheric equivalent with a site-specific marine reservoir correction to provide terrestrial calibration from 12.4–26.0 cal kyr B P. A substantial enhancement relative to IntCal98 is the introduction of a coherent statistical approach based on a random walk model, which takes into account the uncertainty in both the calendar age and the 14C age to calculate the underlying calibration curve (Buck and Blackwell, this issue). The tree-ring data sets, sources of uncertainty, and regional offsets are discussed here. The marine data sets and calibration curve for marine samples from the surface mixed layer (Marine04) are discussed in brief, but details are presented in Hughen et al. (this issue a). We do not make a recommendation for calibration beyond 26 cal kyr BP at this time; however, potential calibration data sets are compared in another paper (van der Plicht et al., this issue).