This dataset includes a high-resolution gridded model hindcast simulation of the Antarctic Peninsula during the period 1998-2017, produced using the Met Office Unified Model (UM).

Variables included in the dataset include near-surface meteorological variables like temperature and relative humidity, atmospheric profiles such as winds and humidity on pressure levels, cloud properties such as liquid/ice water paths, surface energy balance terms such as radiative and turbulent fluxes and surface fields such as surface meltwater production. All variables are outputted at 3- or 6-hourly intervals. Variables are separated into individual netCDF files, which are either two dimensional (for example surface for near-surface meteorological fields) or three dimensional (for example atmospheric profiles), over time.

The region covered is the central and northern Antarctic Peninsula, centred on the Larsen C ice shelf. The simulations are gridded on rotated pole coordinates and cover the period 01-01-1998 00:00 UTC to31-12-2017 23:59 UTC.

A dynamically downscaled regional (limited area) version of the UM is run in atmosphere-only mode at 4.0 km horizontal grid spacing, with 70 vertical levels and a 100 second time step for the inner domain. The model is re-initialised from ERA-Interim reanalysis data every 12 hours, and the time series is produced by concatenating the t+12 hour to t+24 hr segments of each integration into a continuous time series. Specifics of the model configurations and parameterisations used to produce the simulations are documented in Gilbert et al. (2020) (doi: 10.1002/qj.3753).

These simulations were produced as part of the doctoral work of E. Gilbert, and was supported by the Natural Environment Research Council through the EnvEast Doctoral Training Partnership (grant number NE/L002582/1). E. Gilbert also acknowledges the use of the MONSooN system, a collaborative facility supplied under the Joint Weather and Climate Research Programme, a strategic partnership between the Met Office and the Natural Environment Research Council.