Color data, carbon and oxygen isotopes, and offsets from BBCP Polecat Bench cores 2A and 2B

The Earth's climate abruptly warmed by 5-8 °C during the Palaeocene-Eocene thermal maximum (PETM), about 55.5 million years ago**1,2. This warming was associated with a massive addition of carbon to the ocean-atmosphere system, but estimates of the Earth systemresponse to this perturbation are complicated by widely varying estimates of the duration of carbon release, which range from less than a year to tens of thousands of years. In addition the source of the carbon, and whether it was released as a single injection or in several pulses, remains the subject of debate**2-4. Here we present a new high-resolution carbon isotope record from terrestrial deposits in the Bighorn Basin (Wyoming, USA) spanning the PETM, and interpret the record using a carbon-cycle boxmodel of the ocean-atmosphere-biosphere system.Our record shows that the beginning of the PETMis characterized by not one but two distinct carbon release events, separated by a recovery to background values. To reproduce this pattern, our model requires two discrete pulses of carbon released directly to the atmosphere, at average rates exceeding 0.9 Pg C yr**-1, with the first pulse lasting fewer than 2,000 years.

Calanoid copepod abundances in the northern Benguela Current System from 2009 and 2010

Abundance data of copepods were derived from vertical Multinet hauls at 10 stations, carried out in the northern Benguela upwelling system in December 2009 (FRS Africana) and September/October 2010 (RRS Discovery). Three transects along ~ 17°S, 19°S and 23°S with three stations each (neritic, shelf break, oceanic) and one station at 21°S were analysed for copepod abundance. Maximum sampling depth was either close to the seafloor (neritic and shelf break stations) or 700 m (2009) and 1000 m (2010) for the oceanic stations. Calanoid copepod species and stages were identified and enumerated separately. Adult females, males and copepodite stage 5 (C5) (in case of C. carinatus and N. minor) were included in the abundance calculations. Abundance is expressed as number of individuals per m**3, calculated from the volume of water filtered (calibrated flowmeter, Hydro-Bios) and the maximum sampling depth at each station.

Depth integrated nitrogen fixation from the northern Benguela upwelling system during Maria S. Merain cruise MSM18/5

Nitrogen fixation data from the cruise number MSM18/5 with research vessel "Maria S. Merian" from 22.08.-20.09.2011 (from Walvis Bay to Walvis Bay) in front of Angola and northern Namibia. Samples taken by CTD- rosette sampler from different depths and incubated in glass bottles (535 ml) at light intensities that resemble the in situ light intensities of the sampling depth after 15N2 gas was injected to the sample. After the incubation time of 6 hours, the complete bottle content was filtered onto a pre-combusted Whatman GF/F filter. Filters were frozen, transported to the institute on dry ice and measured in a mass spectrometer for Delta 15N. The principle of the method was described by Montoya et al. (1996) and calculation was done according to their spread sheet. From the data of the single depths, the nitrogen fixation per square meter within the upper 40 m of the water column was calculated. The methods are described in detail in a paper submitted by Wasmund et al. in 2014 to be printed in 2015. Some results are surprisingly below zero. This occurs if the Delta 15N of the blank is higher than the measurement after incubation. It indicates that no nitrogen fixation occurred. Due to natural variability, the variability of the nitrogen fixation data is high. In an overall estimate, also over several cruises, negative and positive values compensate more or less, suggesting that nitrogen fixation is insignificant in the waters in front of northern Namibia and southern Angola.

Ichthyoplankton density and zooplankton biomass in the Benguela upwelling system during MSM17/3 in January and February 2011

Ichthyoplankton density (fish eggs and larvae) and bulk zooplankton biomass in January/February 2011 were determined for 38 stations in the northern Benguela upwelling system, based on oblique Multinet hauls during the FS Maria S. Merian MSM17/3 cruise. A HYDROBIOS Multinet, type Midi (0.25 m**2 mouth area) was equipped with five nets of 500 µm-mesh size, temperature and oxygen probes, and an inner and outer flow meter to monitor the net's trajectory (for volume filtered calculations) as well as net clogging. The Multinet was handled over the side, towed horizontally at 2 knots. Winch speed when fearing was 0.5 or 0.3 m/s, heaving velocity 0.2 - 0.3 m/s. The Multinet was towed obliquely at 38 stations sampling the upper 200 m of the water column, which were divided into five different depth strata after inspection of temperature and oxygen concentration depth profiles. Ichthyoplankton densities and zooplankton biomass were calculated for each depth stratum (=single net) from total abundance and the volume of water filtered [individuals per m**3 and g wet weight per m**3, respectively]. In addition, densities and biomass were integrated over the area for each station [individuals per m**2], as sum of calculations for each net: Sum ([individuals per m**3]*Delta (depth bot[m]-depth top [m]).

Ichthyoplankton density and zooplankton biomass in the Benguela upwelling system during MSM19/1b in October 2011

Ichthyoplankton density (fish eggs and larvae) and bulk zooplankton biomass in October 2011 were determined for 22 stations in the northern Benguela upwelling system, based on oblique Multinet hauls during the FS Maria S. Merian MSM19/1b cruise. A HYDROBIOS Multinet, type Midi (0.25 m**2 mouth area) was equipped with five nets of 500 µm-mesh size, temperature and oxygen probes, and an inner and outer flow meter to monitor the net's trajectory (for volume filtered calculations) as well as net clogging. The Multinet was handled over the side, towed horizontally at 2 knots. Winch speed when fearing was 0.5 or 0.3 m/s, heaving velocity 0.2 - 0.3 m/s. The Multinet was towed obliquely at 22 stations sampling the upper 200 m of the water column, which were divided into five different depth strata after inspection of temperature and oxygen concentration depth profiles. Ichthyoplankton densities and zooplankton biomass were calculated for each depth stratum (=single net) from total abundance and the volume of water filtered [individuals per m**3 and g wet weight per m**3, respectively]. In addition, densities and biomass were integrated over the area for each station [individuals per m**2], as sum of calculations for each net: Sum ([individuals per m**3]*Delta (depth bot[m]-depth top [m]).