Location, location, location

30 January 2018 | Richard Burness

When it comes to the analysis of records in the database there are several parameters that are critical to the integrity and value of each record. What species is it? How many were there? Were they calling, singing, nesting? Who saw it? Where is it? The last question seems simple and obvious. However, there are a surprising number of database errors attributable to spatial misplacement which lessen the value of the records in question.

We are requesting that when you submit your records the site names you use are as precise as possible. The ideal record would have a site name taken from an OS map near to the location accompanied by a six-figure grid reference giving the exact location of the bird. Birds reported from popular sites, such as Lower Farm GPs or Hosehill LNR will automatically have a four-figure grid reference allocated by the database but a six-figure reference will still be helpful. Town or village names should only be used for records in the urban area, for example "Thatcham" is not an alternative for "Thatcham Marsh" or "Thatcham GPs". Vague site definitions cause problems. "The Ridgeway", "The K & A Canal" or "Drift Road" reduce the worth of a record because each of these "sites" stretch many miles and cross county boundaries. Of course, the ideal isn't always possible. If you are submitting records electronically, either on an Excel spreadsheet or through Birdtrack, then you can use the notes and remarks columns to qualify the location. It is also worth noting that many smart phone map apps enable you to read out grid references or GPS co-ordinates.

Some of the database problems are caused by duplication of entry. These occur when a site has several different names. It is impossible to eliminate duplicates altogether but using the name in commonest usage would help. (As an example, observers reporting the same bird as being at Eversley GPs, Moor Green Lakes or Yately GPs will find their observations entered as three separate individual birds in the database.)

Finally make sure that the bird is in Berkshire! You'd be surprised at how many records for Surrey, Oxon and Bucks end up in our database. The main problem occurs with records entered through Birdtrack, which resolves records to 10km squares. (In a recent case birds seen at Greys Court in Oxfordshire were allocated to Greys Court in Reading.) Again, the answer is to be as precise as possible.

(As a coda to my previous piece about the work of the BRC, the secretary has noted that there are an increasing number records from single observers or single groups reporting scarce species without providing the supporting evidence. In these circumstances the BRC will, reluctantly, have no option but to reject the records.)