Johan Nyberg

Low Atlantic hurricane activity in the 1970s and 1980s compared to the past 270 years

Hurricane activity in the North Atlantic Ocean has increased significantly since 1995 (refs 1, 2). This trend has been attributed to both anthropogenically induced climate change and natural variability, but the primary cause remains uncertain. Changes in the frequency and intensity of hurricanes in the past can provide insights into the factors that influence hurricane activity, but reliable observations of hurricane activity in the North Atlantic only cover the past few decades. Here we construct a record of the frequency of major Atlantic hurricanes over the past 270 years using proxy records of vertical wind shear and sea surface temperature (the main controls on the formation of major hurricanes in this region) from corals and a marine sediment core. The record indicates that the average frequency of major hurricanes decreased gradually from the 1760s until the early 1990s, reaching anomalously low values during the 1970s and 1980s. Furthermore, the phase of enhanced hurricane activity since 1995 is not unusual compared to other periods of high hurricane activity in the record and thus appears to represent a recovery to normal hurricane activity, rather than a direct response to increasing sea surface temperature. Comparison of the record with a reconstruction of vertical wind shear indicates that variability in this parameter primarily controlled the frequency of major hurricanes in the Atlantic over the past 270 years, suggesting that changes in the magnitude of vertical wind shear will have a significant influence on future hurricane activity.

Comparison of statistical and artificial neural network techniques for estimating past sea surface temperatures from planktonic foraminifer census data

We present the first detailed and rigorous comparison of six different computational techniques used to reconstruct sea surface temperatures (SST) from planktonic foraminifer census data. These include the Imbrie-Kipp transfer functions (IKTF), the modern analog technique (MAT), the modern analog technique with similarity index (SIMMAX), the revised analog method (RAM), and, for the first time, a set of back propagation artificial neural networks (ANN) trained on a large faunal data set, including a modification where geographical information was added among the input variables (ANND). By training the techniques on an identical database, we were able to explore the differences in SST reconstructions resulting solely from the use of different mathematical methods. The comparison indicates that while the IKTF technique consistently shows the worst performance, ANN and RAM perform slightly better than MAT and that the inclusion of the geographical information into the training database (SIMMAX and ANND) further improves the accuracy of modern SST estimates. However, when applied to an independent validation data set and an additional fossil data set, the results did not conform to this ranking. The largest differences in the reconstructed SST values occurred between groups of techniques with different approaches to SST reconstruction; that is, ANN and ANND produced SST reconstructions significantly different from those produced by RAM, SIMMAX, and MAT. The application of the various techniques to the validation data set, which allowed comparison of SST reconstructions with instrumental records, suggests that artificial neural networks might provide better paleo-SST estimates than the other techniques.

DROUGHT AND THE MAYA COLLAPSE

Abstract Between a.d. 760 and 930, millions of Maya disappeared from the Earth. We examine changes in the physical environment in which the Maya lived. The ice-core evidence from Greenland indicates that around the time of the Maya Collapse, a minimum in solar insolation and a low in solar activity occurred, accompanied by severe cold and dryness over Greenland, indicating hemispheric climatic conditions propitious for drought in the Maya Lowlands. In the northeastern Caribbean, sea-surface salinity (SSS) was lowered. The most severe drought of the past 7,000 years devastated the Yucatan Peninsula. Large Maya cities collapsed in four phases of abandonment spaced about fifty years apart around a.d. 760, 810, 860, and 910. A new core taken from Lake Chichancanab in Quintana Roo shows three peak episodes of brutal drought within a 150- to 200-year drought. A marine core from the Cariaco Basin off Venezuela precisely dates four severe drought episodes to 760, 810, 860, and 910, coincident with the four phases of abandonment of cities. The long-term drought appears to have lasted from 760 to 930 in the Cariaco Basin. The climatic changes were the most drastic the Maya had faced in the preceding 1,500 years and the most severe of the preceding 7,000 years.

Integrated use of LiDAR and multibeam bathymetry reveals onset of ice streaming in the northern Bothnian Sea

Abstract Geomorphological mapping from the new LiDAR (Light Detection and Ranging)-derived digital elevation model for Sweden and a high-resolution multibeam bathymetry data-set for the Gulf of Bothnia reveals a continuous system of glacial landforms crossing the transition between the modern terrestrial and marine environments. A palaeo-ice stream in the northern Bothnian Sea is reconstructed, with an onset tributary over the present-day Ångermanland–Västerbotten coastline. Systematic contrasts in landform morphology and lineation length indicate that this ice stream comprised a relatively narrow (∼40 km) corridor of fast flow, flowing first SW then S, and likely fed by converging flow around the upper Bothnian Sea. The geometry and landform associations of this system imply that ice, at the time period represented here, did not flow across the Gulf of Bothnia: SSE-ward ice flow indicators on the northern Swedish coast do not correspond directly with landform assemblages of the large SE-oriented Finnish deglacial lobes. Instead, we suggest they may contribute to a late-stage fast-flow event to the S and SW. Multibeam bathymetry data offer entirely new access into the rich, landform-scale geomorphological record on the seafloor of the Gulf of Bothnia. The combination of offshore multibeam with the new terrestrial LiDAR data provides unprecedented insight into and renewed understanding of the glacial dynamics of the Bothnian Sea sector of the Fennoscandian Ice Sheet, hitherto interpreted over large areas of unmapped ice sheet bed.

Coral windows onto seasonal climate variability in the northern Caribbean since 1479

Mean surface ocean conditions in the Caribbean were up to ∼2°C cooler than today at times during the Little Ice Age. The seasonal context for such mean state changes is important for determining the mechanisms involved. We reconstructed surface ocean conditions in southwest Puerto Rico at approximately monthly resolution over eight 4–12 year periods during the last ∼520 years to test if the seasonal cycles of temperature or salinity varied with mean state. We carried out paired analyses of Sr/Ca and δ18O for two coral cores. The δ18O data contained clear annual cycles and were significantly correlated to temperature during the 20th century calibration periods (1993–2004 and 1902–1912, r = 0.73). The Sr/Ca data contained high-frequency noise that obscured the seasonal cycles, although the centennial variability matched that of the coral δ18O, indicating a common forcing that is likely temperature. The seasonal coral δ18O amplitude averaged 0.60 ± 0.17‰, with none of the periods significantly different from the most recent. The simplest explanation is that the amplitudes of seasonal seawater δ18O and temperature variations were not different from today. Previous work in the southern Caribbean indicates that the Intertropical Convergence Zone was shifted southward or weaker during the Little Ice Age, and we speculate about how this could occur with no apparent affect on seasonality in the northern Caribbean.

Nyberg et al. reply

Replying to: U. Neu 451, 10.1038/nature06576 (2008).Neu suggests that the reconstruction of Atlantic major hurricane activity (MHA) (that is, frequency) in Nyberg et al. overestimates past MHA because it differs significantly from the known observational records of tropical storms and MHA before 1945 and overestimates the influence of vertical windshear |Vz|.