Methodology:

Breeding counts are made 7 days after 75% of monitored marked nests in Square Pond and Johnson colonies have laid at least one egg. All occupied nests i.e. those with eggs are counted at Johnson and Square Pond on the appropriate day. All nests are checked and all those containing eggs tallied at each colony. Counts are done at Natural Arch and Mountain Cwm within one day of the Johnson Counts. Separate counts are recorded for the Upper and Lower colonies at Natural Arch and Mountain Cwm. Counts are done 3 times until the numbers are similar (within around 5%), or if two people are counting, then until the two counts are within around 5%.

Chicks are counted at all the above colonies 98 days after peak laying (when 75% of monitored nests have laid eggs). Separate counts are made for the Upper and Lower colonies at Natural Arch and Mountain Cwm. Counts are made on the appropriate dates according to the dates used for the breeding counts. Counts at Natural Arch and Mountain Cwm are made within one day of the Johnson counts as for the breeding counts. Counts are done 3 times until numbers are similar (within around 5%)

All data were collected under permit from the Government of South Georgia and the South Sandwich Islands.

All sampling procedures used were regulated by the British Antarctic Survey Ethical Review Committee in collaboration with Cambridge University and the UK Home Office.