Arctic Siberia: Modern pollen dataset and pollen spectra changes over the last 7,000 years from a lacustrine sediment core (southern Taymyr region)

Palaeoecological investigations in the larch forest-tundra ecotone in northern Siberia have the potential to reveal Holocene environmental variations, which likely have consequences for global climate change because of the strong high-latitude feedback mechanisms. A sediment core, collected from a small lake (radius ~100 m), was used to reconstruct the development of the lake and its catchment as well as vegetation and summer temperatures over the last 7100 calibrated years. A multi-proxy approach was taken including pollen and sedimentological analyses. Our data indicate a gradual replacement of open larch forests by tundra with scattered single trees as found today in the vicinity of the lake. An overall trend of cooling summer temperature from a ~2 °C warmer-than-present mid-Holocene summer temperatures until the establishment of modern conditions around 3000 years ago is reconstructed based on a regional pollen-climate transfer function. The inference of regional vegetation changes was compared to local changes in the lake's catchment. An initial small water depression occurred from 7100 to 6500 cal years BP. Afterwards, a small lake formed and deepened, probably due to thermokarst processes. Although the general trends of local and regional environmental change match, the lake catchment changes show higher variability. Furthermore, changes in the lake catchment slightly precede those in the regional vegetation. Both proxies highlight that marked environmental changes occurred in the Siberian forest-tundra ecotone over the course of the Holocene.

Spectral reflectance of ODP Leg 178 holes

The routine use of spectrophotometry on the sediment surfaces of archive halves of each section during the onboard sedimentological core description process is a great stride toward development of real-time noninvasive characterization of deep-sea sediments. Spectral reflectance data have been used so far for mineral composition studies as well as for lithostratigraphic correlation between sites (Balsam and Deaton, 1991; Balsam et al., 1997; Mix et al., 1995; Ortiz et al., 1999). Their results demonstrate that spectrophotometry can estimate CaCO3 content by using the 4.65-, 5.25-, and 5.55-µm wavelength spectrums. A detailed overview of various other noninvasive methods is given in Ortiz and Rack (1999). The purpose of this study is to test whether spectrophotometry in the visible band can be used as a tool to gather further information about grain-size variation, sorting, compaction, and porosity, which are directly linked to the sedimentation process. From remote sensing data analyses, it is known that diffuse spectral reflectance data in the visible band in the wavelength window of 7.0-6.5 µm are sensitive to grain-size variations. It appears that a relationship between grain size and signal absorption exists only in this wavelength window. (e.g., Clark, 1999; Gaffey, 1986; Gaffey et al., 1993). Variations in grain size during a sedimentation process are linked to depositional energy, which affects sorting, compaction, and porosity of sediment deposits. As an example, we study here the spectrophotometric data of the sedimentary sequence of Hole 1098C, which was deposited under widely varying environmental conditions. Alternating turbidite and finely laminated sediments were recovered from Hole 1098C. The turbidites are related to a high depositional energy environment; the finely laminated sediments are related to a low depositional energy environment. Data from Hole 1098C were therefore used to test whether the spectral reflectance data can provide a proxy for these different depositional environments.

(Table 1) Ice flux, polynya area and sea-ice drift speed in the Kara Sea during winter (Jan-Apr) 1997-2001

The Shelf Seas of the Arctic are known for their large sea-ice production. This paper presents a comprehensive view of the Kara Sea sea-ice cover from high-resolution numerical modeling and space-borne microwave radiometry. As given by the latter the average polynya area in the Kara Sea takes a value of 21.2 × 10**3 km**2 ± 9.1 × 10**3 km**2 for winters (Jan.-Apr.) 1996/97 to 2000/01, being as high as 32.0 × 10**3 km**2 in 1999/2000 and below 12 × 10**3 km**2 in 1998/99. Day-to-day variations of the Kara Sea polynya area can be as high as 50 × 10**3 km**2. For the seasons 1996/97 to 2000/01 the modeled cumulative winter ice-volume flux out of the Kara Sea varied between 100 km**3/a and 350 km**3/a. Modeled high (low) ice export coincides with a high (low) average and cumulative polynya area, and with a low (high) sea-ice compactness in the Kara Sea from remote sensing data, and with a high (low) sea-ice drift speed across its northern boundary derived from independent model data for the winters 1996/97 to 2000/01.

Occurrence (presence-only) of living Lophelia pertusa reefs in the Irish continental margin

The present data set was used as a training set for a Habitat Suitability Model. It contains occurrence (presence-only) of living Lophelia pertusa reefs in the Irish continental margin, which were assembled from databases, cruise reports and publications. A total of 4423 records were inspected and quality assessed to ensure that they (1) represented confirmed living L. pertusa reefs (so excluding 2900 records of dead and isolated coral colony records); (2) were derived from sampling equipment that allows for accurate (<200 m) geo-referencing (so excluding 620 records derived mainly from trawling and dredging activities); and (3) were not duplicated. A total of 245 occurrences were retained for the analysis. Coral observations are highly clustered in regions targeted by research expeditions, which might lead to falsely inflated model evaluation measures (Veloz, 2009). Therefore, we coarsened the distribution data by deleting all but one record within grid cells of 0.02° resolution (Davies & Guinotte 2011). The remaining 53 points were subject to a spatial cross-validation process: a random presence point was chosen, grouped with its 12 closest neighbour presence points based on Euclidean distance and withheld from model training. This process was repeated for all records, resulting in 53 replicates of spatially non-overlapping sets of test (n=13) and training (n=40) data. The final 53 occurrence records were used for model training.