Geochemical data has been collected on samples from new exposures of the 1883 deposits, revealed by the 2018 tsunamigenic flank collapse of Anak Krakatau, which provides improved stratigraphic context. Whole-rock data taken by X-ray Florescence shows no systematic stratigraphic correlation. Chemical data for transects across, and spot points on, plagioclase phenocrysts, including some trace element data, all obtained using Electron Probe Microanalysis (EPMA), with Backscatter electron (BSE) images of crystals, obtained using Scanning Electron Microscope, reveal complex zoning profiles. However, chemical data for transects across pyroxene phenocrysts, obtained using EPMA, show this phenocryst phase is largely unzoned. The dataset also includes chemical data for spots on Fe/Ti oxides, included on the rims of pyroxene, and obtained using EPMA. Matrix glass chemistry, obtained via EPMA, shows that the early eruptive ash is more evolved than the pyroclastic material that follows, and that there is a slight overall trend to a more homogenous, less evolved melt composition. The 1883 eruption of Krakatau was a large, cardera-forming eruption that caused approximately 36,000 fatalities. It is also the only eruption of its size to have accompanying written accounts.

This data set includes the original time series collected with broadband and long-period MT instruments during two field seasons in 2016 and 2017 by a team of researchers from the University of Edinburgh, UK and the Institute for Geophysics, Space Science and Astronomy at Addis Ababa university, Ethiopia. For the magnetotelluric stations, processed transfer functions are included in the edi file format. The time series data is provided both in the original raw data format and an ascii version. We provide information on the locations and the processing and include the necessary instrument response functions and metadata to reproduce our results from the raw data. For the TEM recordings, site coordinates and raw data are included in the original format.

Data from laboratory experiments conducted as part of project NE/K011464/1 (associated with NE/K011626/1) Multiscale Impacts of Cyanobacterial Crusts on Landscape stability. Soils were collected from two sites in eastern Australia and transferred to a laboratory at Griffith University, Queensland for conduct of experiments. Soils were A, a sandy loam, and B a loamy fine sand. Trays 120 mm x 1200 mm x 50 mm were filled with untreated soil that contained a natural population of biota. Soils were either used immediately for experiments (physical soil crust only: PC) or were placed in a greenhouse and spray irrigated until a cyanobacterial crust has grown from the natural biota. Growth was for a period of 5 days (SS), c.30 days (MS2) or c.60 days (MS1). Following the growing period (if applicable) trays were placed in a temperature/humidity controlled room at 35° and 30% humidity until soil moisture (measured 5 mm below the surface) was 5%. Trays were then subject to rainfall simulation. Rainfall intensity of 60 mm hr-1 was used and rainfall was applied for 2 minutes (achieving 2 mm application), 8 minutes (achieving 8 mm application) or 15 minutes (achieving 15 mm application). Following rainfall, trays were returned to the temperature/humidity-controlled room under UV lighting until soil moisture at 5 mm below the surface was 5%. A wind tunnel was then placed on top of each tray in turn and a sequential series of wind velocities (5, 7, 8.5, 10, 12 m s-1) applied each for one minute duration. On each tray the five wind velocities were run without saltation providing a cumulative dust flux. For the highest wind speed, an additional simulation run was conducted with the injection of saltation sands. Three replicates of each rainfall treatment were made. Variables measured include photographs, spectral reflectance, surface roughness, fluorescence, penetrometry, chlorophyll content, extracellular polysaccharide content, Carbon, Nitrogen and splash erosion and particle-size analysis (of wind eroded material). Details of rainfall simulator, growth of cyanobacteria, laser soil surface roughness characterisation and wind tunnel design and deployment in Strong et al., 2016; Bullard et al. 2018, 2019. Bullard, J.E., Ockelford, A., Strong, C.L., Aubault, H. 2018a. Impact of multi-day rainfall events on surface roughness and physical crusting of very fine soils. Geoderma, 313, 181-192. doi: 10.1016/j.geoderma.2017.10.038. Bullard, J.E., Ockelford, A., Strong, C.L., Aubault, H. 2018b. Effects of cyanobacterial soil crusts on surface roughness and splash erosion. Journal of Geophysical Research – Biogeosciences. doi: 10.1029/2018. Strong, C.S., Leys, J.F., Raupach, M.R., Bullard, J.E., Aubault, H.A., Butler, H.J., McTainsh, G.H. 2016. Development and testing of a micro wind tunnel for on-site wind erosion simulations. Environmental Fluid Mechanics, 16, 1065-1083.

NERC Grant NE/M011488/1 Electron microprobe analyses of Fe-oxide and Fe-oxyhydroxide phases as elemental percentages per point analysis. The phases were within limonites from Acoje (Philippines), Caldag (Turkey), Nkamouna (Cameroon), Piaui (Brazil) and Shevchenko (Kazakhstan) laterite deposits. The data were acquired during the NERC SoS Minerals CoG3 project between 2015 and 2018 using a Cameca SX100 electron microprobe at the Natural History Museum, London, UK. Point analyses were performed on samples set within epoxy resin blocks, polished and coated with carbon. All elements were analysed using wavelength dispersive X-ray spectrometers. These data were used to identify the Co and Ni bearing host minerals within each natural resource and to assess the amount and variability of these elements within specific Fe-oxide or Fe-oxyhydroxide phases. This may be useful within the mining sector, resource assessment, processing or prospecting, geo- or material scientists and processing engineers / metallurgists. The data were acquired in the Core Research Laboratories, Natural History Museum by the NHM CoG3 team. NERC grant: CoG3: The geology, geometallurgy and geomicrobiology of cobalt resources leading to new product streams

The Molloy Hole was surveyed with a Kongsberg EM 124 gondola-mounted to the hull of the 225-foot DSSV Pressure Drop. The survey was conducted over the course of three days – August 24-26, 2019. The data meet the requirements for IHO Order 1 standards.

The RiftVolc microgravity network was comprised of a total of 4 benchmarks including a reference benchmark. Benchmark locations, observed gravity changes, dg14 -16, from 2014-2016, corresponding vertical deformation, Uz, free-air effect, and resultant residual gravity changes gr of the microgravity and GNSS network at Corbetti.

Two sediment depth cores were collected from a mud sediment patch at the sea floor of the Irish Sea. Cores were collected by the University of Manchester. Cores were sliced at 1 cm intervals from 0 - 10 cm, and at 2 cm intervals thereafter. Slicing was performed in an anaerobic bag. Samples were transferred to Newcastle University for DNA extraction. A total of 21 samples were extracted for core 1, and 23 samples extracted for core 2.

Sediment data and Nd data from fish teeth in samples from IODP Site 1490. NERC grant The Late Miocene Climate Enigma: Insights from Expedition 363.

Thermogravimetric Analysis profiles as xy datasets of limonites from Acoje (Philippines), Caldag (Turkey), Piaui (Brazil) and Shevchenko (Kazakhstan) laterite deposits. The data were acquired during the NERC SoS Minerals CoG3 project between 2015 and 2018 using a TA Instruments Thermogravimetric Analyzer (TGA) instrument at the Natural History Museum, London, UK. Powdered samples were loaded into the TGA and heated at 10 degrees C per minute up to 800 degrees C under a flowing N2 atmosphere. Information regarding the type and proportion of hydrous or volatile rich mineral phases can be obtained from the decomposition profile. This may be useful within the mining sector, resource assessment, processing or prospecting, geo- or material scientists and processing engineers / metallurgists. The data were acquired in the Earth Sciences Department, Natural History Museum by the NHM CoG3 team. NERC grant: CoG3: The geology, geometallurgy and geomicrobiology of cobalt resources leading to new product streams

Data from laboratory experiments conducted as part of project NE/K011464/1 (associated with NE/K011626/1) Multiscale Impacts of Cyanobacterial Crusts on Landscape stability. Soils were collected from two sites in eastern Australia and transferred to a laboratory at Griffith University, Queensland for conduct of experiments. Soils were A, a sandy loam, and B a loamy fine sand. Trays 120 mm x 1200 mm x 50 mm were filled with untreated soil that contained a natural population of biota. Soils were either used immediately for experiments (physical soil crust only: PC) or were placed in a greenhouse and spray irrigated until a cyanobacterial crust has grown from the natural biota. Growth was for a period of 5 days (SS), c.30 days (MS2) or c.60 days (MS1). Following the growing period (if applicable) trays were placed in a temperature/humidity controlled room at 35° and 30% humidity until soil moisture (measured 5 mm below the surface) was 5%. Trays were then subject to rainfall simulation. Rainfall intensity of 60 mm hr-1 was used and rainfall was applied for 2 minutes (achieving 2 mm application), 8 minutes (achieving 8 mm application) or 15 minutes (achieving 15 mm application). Following rainfall, trays were returned to the temperature/humidity-controlled room under UV lighting until soil moisture at 5 mm below the surface was 5%. A wind tunnel was then placed on top of each tray in turn and a sequential series of wind velocities (5, 7, 8.5, 10, 12 m s-1) applied each for one minute duration. On each tray the five wind velocities were run without saltation providing a cumulative dust flux. For the highest wind speed, an additional simulation run was conducted with the injection of saltation sands. Three replicates of each rainfall treatment were made. Variables measured include photographs, spectral reflectance, surface roughness, fluorescence, penetrometry, chlorophyll content, extracellular polysaccharide content, Carbon, Nitrogen and splash erosion and particle-size analysis (of wind eroded material). Details of rainfall simulator, growth of cyanobacteria, laser soil surface roughness characterisation and wind tunnel design and deployment in Strong et al., 2016; Bullard et al. 2018, 2019. Bullard, J.E., Ockelford, A., Strong, C.L., Aubault, H. 2018a. Impact of multi-day rainfall events on surface roughness and physical crusting of very fine soils. Geoderma, 313, 181-192. doi: 10.1016/j.geoderma.2017.10.038. Bullard, J.E., Ockelford, A., Strong, C.L., Aubault, H. 2018b. Effects of cyanobacterial soil crusts on surface roughness and splash erosion. Journal of Geophysical Research – Biogeosciences. doi: 10.1029/2018. Strong, C.S., Leys, J.F., Raupach, M.R., Bullard, J.E., Aubault, H.A., Butler, H.J., McTainsh, G.H. 2016. Development and testing of a micro wind tunnel for on-site wind erosion simulations. Environmental Fluid Mechanics, 16, 1065-1083.