U12B-01 INVITED
Solar Variability, Volcanic Eruptions, and Surface Temperature Over the Past Two Millennia
Global and hemispheric mean surface temperatures vary in response to forcings external to the climate system that affect the Earth's energy balance. For the last 2000 years, the dominant natural forcings are probably aerosols injected high into the atmosphere by explosive volcanic eruptions and fluctuations in solar radiation reaching the Earth associated with solar activity. The small variations in greenhouse gas concentrations prior to the industrial revolution are probably some mix between an internal, natural feedback of the carbon cycle to climate fluctuations and an external, direct effect of human activity. Climate model simulations indicate that solar and volcanic forcings together could have produced the periods of relative warmth and cool conditions during the preindustrial part of the last 1000 years. Reconstructions of these natural forcings during the past 2 millennia are largely based on proxy indicators from ice cores and tree rings that are calibrated against (and extrapolated from) recent instrumental data. The interpretations of these proxies involve numerous uncertainties, particularly regarding the magnitude connected to the range of variations. Recent satellite measurements suggest that the magnitude of millennial variations in solar variability may need to be reassessed. Quantifying the natural solar and volcanic forcings of the past 2 millennia in conjunction with the natural variability of surface temperatures can provide better understanding of the climate sensitivity indicated in climate models.
U12B-02
Response of Late Holocene Nordic Sea temperatures to solar forcing
We report on a new, annually- to decadally-resolved oxygen isotope paleotemperature series developed in sediment cores (P1003MC and P1003SC) raised from a region of rapid sedimentation and strong Atlantic water inflow in the SE Norwegian Sea. A chronology of the last 8000 years has been established based on identification of historic Icelandic tephras, lead dating and 90 AMS radiocarbon dates. Identification of radiocarbon plateaus within the sediment record has allowed us to constrain the marine reservoir age of radiocarbon and to estimate calendar ages via wiggle matching against the 14C age v. calendar age calibration relationship in tree rings, greatly improving the chronology. Previous studies have demonstrated strong correlation between 18O/16O in Neogloboquadrina pachyderma (right coiling form) and the local instrumental record of summer air temperature back to ~1870 AD (Berstad et al., 2003). Over the last 1000 yr, estimated sea surface and coastal air temperatures varied by 1-2 �C and appear to co-vary (r = ~ 0.6) with observational (group sun spot number and galactic cosmic ray flux) and proxy (14C and 10Be) records of solar activity. The relation of temperature and solar activity is also directly evident from the location of major 14C plateaus (3 in the last 700 years) relative to shifts in oxygen isotope composition within the sediments. The new isotope series is also similar to various reconstructions of Northern Hemisphere temperature, but has significantly larger amplitude throughout the last 1000 years and obvious differences in structure at 1800-1900 AD, and 1500-1600 AD. These differences suggest that solar variability may have a larger impact on atmosphere/ocean dynamics which redistribute ocean heat within the hemisphere than it does on the hemispheric energy balance as a whole (consistent with generally low estimates of Total Solar Irradiance change on these timescales). Lastly, we suggest a strategy of placing records of Atlantic SST and salinity on a common timescale using 14C plateaus as a means of testing mechanisms of heat and freshwater redistribution within the basin on centennial timescales.
U12B-03
A 1400-Year Record of Sea-Surface Temperature Variability From the Northern Gulf of Mexico
A continuous, decadal-scale resolution sea-surface temperature (SST) record of the past 1400 years in the northern Gulf of Mexico (GOM) was constructed from a box core recovered from the Pigmy Basin. SST was calculated from Mg/Ca measurements on the white variety of the planktic foraminifer Globigerinoides ruber. Age control for box core PBBC-1 was based on seven AMS $^{14}$C dates, and extended to a zero-age core-top. Two multi-decadal intervals of sustained high Mg/Ca values indicate GOM SSTs were as warm or warmer than core- top SSTs between 1000 and 1400 yrs BP. Foraminiferal Mg/Ca values during the coolest interval of the Little Ice Age (ca. 250 yrs BP) indicate that SST was 2 - 2.5 $\deg$C below modern SST. The total amplitude of SST variability in the GOM over the past 1400 years is 2.5-3 $\deg$C. Four minima in the Mg/Ca record between 900 and 250 yrs BP correspond with the Maunder, Sp�rer, Wolf and Oort sunspot minima, suggesting a link between solar insolation and SST variability in the GOM. The amplitude of temperature variability observed in the Pigmy Basin record is consistent with other temperature proxy records from the western subtropical Atlantic, and the pattern of variability closely corresponds to the tree-ring based Northern Hemisphere reconstruction of Esper et al. (2002).
U12B-04
Sea Surface Temperature and Salinity Variability in the Indo-Pacific Warm Pool During the Last 2000 Years
The Indo-Pacific Warm Pool is one of the warmest regions in the global ocean and is a significant source of interannual to decadal climate variability through the El Ni�o- Southern Oscillation and Pacific Decadal Oscillation. Here we present sea surface temperature (SST) and salinity records for the last 2000 years derived from Mg/Ca ratios and $\delta ^{18}$ O values of planktonic foraminifera from two sediment cores collected from the Makassar Strait, Indonesia. High sedimentation rates (~ 40cm 10$^{-3}$ years) in this region allow us to resolve decadal to centennial scale variability, as well as longer term trends during the last two millennia. The period from 2000 to 1000 Cal YBP was marked by a slight increase in average SST of < 0.5$\deg$C in the southern end of the Strait and more pronounced in the northern end of the Strait (~1$\deg$C). From about 1000 to 150 Cal YBP SST decreased by ~1$\deg$C and salinity decreased by ~1 throughout the Strait. Several global scale climate fluctuations are apparent in these records. The warmest temperatures and highest salinities occur during the Medieval Warm Period (1000 to 700 Cal YBP). Average temperatures during this time were ~29.5$\deg$C with strong decadal scale fluctuations. This period is followed by a sharp decline in SST and salinity, with the lowest SSTs (~28.5$\deg$C) occurring during the Little Ice Age (400 to 200 Cal YBP). Conditions were also significantly drier in this region during the Little Ice Age and this is attributed to a southern displacement of the Inter-Tropical Convergence Zone.
U12B-05 INVITED
Glaciologic information on the uniqueness -- or not -- of recent climate change.
Recent controversy over the quality of large-scale surface temperature reconstructions became prominent because some opine that warming of recent decades is part of natural variability and in particular a return to pre-Little Ice Age conditions of Medieval warm climate. In this context I review evidence from ice core isotopic and borehole records, and glacier retreat histories. These provide information separate from statistical dendroclimatologic reconstructions. Emphasis will be on Tropical climate history, coherence of climate changes between polar regions, and major sources of ignorance about both.
U12B-06
Surface Temperature Reconstructions for the Last Five Centuries From Borehole Temperature Profiles
Temperature profiles in terrestrial boreholes contain valuable information about surface temperature changes over the past five centuries. Geothermal analyses of climate change are particularly valuable because deep borehole temperatures (1) are a direct rather than proxy measure of surface temperature change, (2) need no calibration to the surface air temperature record, (3) respond to annual rather than seasonal surface temperatures, and (4) capture long-period (decadal to centennial) surface temperature changes. Several hundred boreholes distributed over all continents are available for analysis, although the locations are concentrated in the mid latitudes of the northern hemisphere. Independent analyses of the global borehole temperature database by three research groups using different data reduction and inversion techniques (Huang et al., 2000; Harris and Chapman, 2001; Beltrami 2002) arrive at similar although not identical reconstructions. For the period 1850 -2000, borehole reconstructions suggest a northern hemisphere warming of about 0.7 $\pm$0.2 $^{o}$C, consistent with the instrumental record; some areas at high latitudes show much greater warming. An additional 0.3 $^{o}$C of warming between 1500 and 1850 is indicated by some analyses. Although early multi-proxy reconstructions departed significantly from the borehole temperature reconstructions, more recent multi-proxy analyses designed to extract low-frequency information are now in essential agreement with the borehole reconstructions to about 1600 AD. Remaining differences between multi-proxy and borehole reconstructions can perhaps be explained by different spatial sampling between the various data sets and seasonal responses of the proxies. Thus a robust, baseline temperature (ca 1600-1800) seems to be emerging against which 20th and 21st century warming can be referenced.
U12B-07 INVITED
Underwater Dendrochronology of the Sierra Nevada: Testing the Medieval Mega-Drought Hypothesis
As stated in the NAS STR Report, "regional and large-scale reconstructions of changes in other climatic variables, such as precipitation, over the last 2,000 years would provide a valuable complement to those made for temperature." In this context, we focus on the 'Medieval Mega-drought Hypothesis', which is based on radiocarbon dating of dead trees and stumps found underwater in Sierra Nevada lakes and streambeds, and states that century-long dry periods caused lakes to retreat and streams to dry up, with the most recent mega- droughts happening during medieval times. While several paleoclimatic records support this hypothesis, some do not, and the possibility exists that geomorphic processes, such as landslides caused by seismic events, were responsible for the presence of trees and stumps under current bodies of water. Given the relevance of this hypothesis, not only for sustainable water management but also for social stability and security, it is necessary to test it beyond reasonable doubt. One way to do so is by measuring the location, orientation, and time of origin of underwater trees, to determine if they were transported or grew in situ. For example, during 2005 wood samples were retrieved from submerged trees at Fallen Leaf Lake, California. The trees had been previously located and documented using an ROV that can obtain high resolution color video, and collect small surface samples using a gripper, down to a water depth of about 150 m. For tree-ring dating, a reference chronology from AD 543 to 2003 was developed using live and dead western juniper trees located near the lake. One underwater sample, i.e. a branch cross section that included 69 rings, was then dated to AD 1085- 1153. This shows that it is feasible to obtain calendar dates and continuous ring-width series from underwater trees in the Sierra Nevada. Submerged trees in Fallen Leaf Lake were mapped in summer 2006 using an EdgeTech 4200 side-scan system capable of decimeter resolution. The 5 km$^{2}$ side-scan survey required about two days of survey time, providing rapid identification of the underwater targets. Side-scan sonar imagery was complemented with sub- bottom seismic CHIRP profiling to assess fault movement and landslide formation. Initial evaluation of these data does not suggest a landslide origin for submerged trees in Fallen Leaf Lake. However, sub-bottom mapping of the West Tahoe fault underlying the southernmost end of Fallen Leaf Lake, combined with evidence that these trees are at a depth equivalent to the modern level of Lake Tahoe, may suggest a linkage between Fallen Leaf Lake levels and earthquake-generated permeability along the West Tahoe fault. These preliminary results indicate that the question of climatic vs. non-climatic origin of submerged trees is best answered by mapping and dating them in more than one lake. As tree-ring samples are collected and entered into a master chronology, they could also yield a continuous, annually resolved record over several millennia, because some submerged trees have already been radiocarbon dated to the mid-Holocene. It is by collating a myriad of wood samples buried in streambeds and lakes that the longest tree-ring chronologies, now going back 12,460 years, have been developed in central Europe.
U12B-08
Climate variability in North America during the past 2,000 years using pollen data
The debate concerning the causes of the global temperature increase of the past century results in part from the difficulty in establishing the magnitude of surface temperature change that can be attributed to natural climate variability and that caused by human impact through greenhouse gas emissions. Recent high-resolution paleoclimate reconstructions of the past two millennia using tree-ring and other proxies correlated well in structure and in timing, however, the magnitude of temperature change in these studies divert from one another. This remains a problem that has plagued these studies and the current debate over how much of the recent surface temperature change can be attributed to human impact. We present a new high-resolution summer temperature reconstruction for North America using an extensive network of pollen records. When compared with northern hemispheric temperature anomaly curves based on tree-ring and multi-proxy reconstructions of the past 2,000 years, the results are remarkably similar, in spite of the different methods and proxies employed in these studies. The magnitude of the temperature changes during the Medieval Warm Period and Little Ice Age are comparable to those derived from other proxy records. Regional temperature reconstructions further document the nature of climate changes across North America. Our results show that these recent climate variations between slightly warmer and cooler conditions are part of an ongoing periodic millennial-scale variability that has been operating throughout the Holocene, with irregularly occurring oscillations ~1100 years apart. Quantification of these millennial-scale variations using pollen series are on the order of +/- 0.2�C which is consistent with the climate variations of the past 2,000 years reconstructed using tree-ring series. We hypothesize that the recent warming of ~ +0.6�C of the past century is unprecedented during at least the past 8,000 years.