Greenshank national survey in the UK, 1995

Quality Control

These data have been collected by trained field workers and are of good quality. All the data have been checked for sensitivities and geographical/typographical errors.

Method steps

1. The breeding range of the Greenshank was defined using records from the 1968-72 and 88-91 breeding birds atlases and additional records from 1968-94 from SSDB (sensitive species database). Evidence for breeding was noted in 332 10-km squares and this formed the breeding range for Greenshanks. The squares were then stratified into high and low categories using abundance data, the 10-km were split into 5-km squares giving a total of 348 and 980 in the high and low stratum respectively. 49 low and 52 high stratum squares were then selected at random to give a sampling intensity for the whole range of around 8%.
   
   Each 5km square was sampled first between the 10th April and the 25 May and then again between 26 May and 10 July. The dates were chosen as a compromise between the typical arrival time and departure time of Greenshanks to their territories. Surveys took place at any time before 18.00GMT and were not undertaken during times of low visibility, high wind or continuous rain. Sampling effort was concentrated in accordance with Greenshank ecology so certain areas were excluded (areas above 800m, slopes steeper than 13%, built up areas, enclosed farmland, dense native or planted forest of thicket stage or older, sea cliffs, the sea below low tide and off-shore islands smaller than 25ha).
   
   'Wet' (lochs, pools systems, rivers, wet bogs and coastal habitats not listed above) habitats were surveyed in their entirety and in 'dry' (peatland without standing water, dry moorland, open or recently planted forests) habitats surveyors passed within 500m of all areas. Any Greenshanks were recorded on a map and the territories in this dataset were then derived from this.

Approximately 8% of the breeding range of the Greenshank was surveyed. The breeding range is entirely in Scotland and was divided up into 332 5 km squares. These squares were stratified into high and low quality and a random sample of the both the high and low quality squares were surveyed (49 low and 52 high).