Atlas 2020, c'est tout finis

As you may have noticed, we are no longer in 2019, meaning that my favourite blog post topic, the BSBI's Atlas 2020 project, is finished. I have been meaning to write something to mark the occasion and celebrate everyone's hard work, but appallingly have been preoccupied with bryology and numerous other things this winter. However, I currently find myself becalmed in an Essex hotel, so here are some concluding thoughts on Atlas 2020 in VC23.

First, it has a been a great pleasure to receive everyone's records over the last three and a bit years. Seeing what local botanists have been up to has been very interesting and enjoyable, particularly being able to share in the excitement of unusual discoveries. I have had records from a relatively small number of regular contributors, either individuals or group recorders, some of whom have produced a considerable volume of records, as well as less frequent recorders. Many of the records that have come through me have been by email, but over the summer of last year the BSBI Distribution Database made friends with the iRecord database which gave me access to the 12,000 or so records from that source. Many of the iRecord records have been collected by a greater diversity of people than belong to the quite small circle of botanists associated with our local flora groups and the BSBI.

It is invidious of me to mention names, but special thanks must go to Tim Harrison, Jonathan Shanklin and Pete Stroh, none of whom live in the county (Jonathan and Pete live as far away as Cambridge and Peterborough, respectively) but who nevertheless spent a lot of time thoroughly surveying under-recorded areas of VC23. From the locals, we have also had regular and significant contributions from Sally Abbey, Fay Banks, Nick Barber, Brenda Betteridge, Susan Erskine, Renée Grayer, Sue Helm, Roger Heath-Brown, Frank Hunt, Clare Malonelee, Oli Pescott, Sally Rankin, Ruth Ripley, Barbara Spence and Frances Watkins. Judy Webb has been very helpful in sending me lists from her prodigious site monitoring work. Of course, as described in a previous blog, a huge contribution to the county's vascular plant data holding comes to me via my data sharing agreement with TVERC. I am very grateful to their data officer Ellen Lee for organising this, as well as to the many recorders who submit their records to TVERC, whom I am sure would be pleased to know that their records contribute to national recording projects like Atlas 2020.

Atlas records per year, excluding duplicates. Red lines are yearly number of records, on the left vertical axis, and blue lines the cumulative total, on the right. Solid and dashed lines are numbers of records with and without TVERC records.

Overall, the records gathered between 1st January 2000 and 31st December 2019 number a little over 320,800 (excluding duplicates). You can see how the records were accumulated in the graph to the left. This total is similar to most other English vice counties, being a little more than the median (~305,200) and less than the mean (~336,100). I have verified all of the Atlas records for VC23, confirming about 330,000 records (including duplicates). Questions marks remain over around 1,200, but thank you if you helped me resolve any queries.

I have shown the split between records from TVERC and other sources to highlight the importance of our data sharing agreement, which makes up more than half of the total records for Atlas 2020. This agreement with TVERC has provided a steady stream of records each year, particularly from important privately owned Local Wildlife Sites, contrasting with the more stop-start contributions direct to the BSBI. What is evident from the graph is the late start to systematic recording of the county for the Atlas project, which began I believe when Sue Helm took on the county recorder job and Atlas recording meetings were first organised. Many thanks to Sue for her hard work during her three year tenure as recorder. The spikes and troughs in the later Atlas period reflect the comings and goings of the non-VC23 based botanists thanked above.

In terms of spatial coverage and success, the county can be pleased with its Atlas dataset. This is summarised in the series of maps shown, which are stills from my interactive Atlas webmap. The project was never supposed to be a complete survey of every tetrad (2km square) in the county, but a repeat of previous hectad (10km square) national atlas surveys, based on a sample of tetrads. The re-recording rate (the coloured proportion of grid squares in the maps) is the crucial statistic for measuring progress — while there have definitely been changes to our flora, most species will still be present in most squares, so the higher the re-recording rate the more thorough and representative the sample is. The BSBI advised that a hectad re-recording rate of 70-80% should be aimed for, and this has been met for many hectads, and just about all compared to the recording period for the New Atlas (1987-1999). On the other hand, since we have no way of measuring abundance, records from too thorough a search might belie real declines, with for instance a formerly common plant reduced to a single plant both registering as present in the Atlas.

The above picture is somewhat complicated by recording effort over different periods and across taxa. The most intensive and extensive period of recording in the county was for the 1997 flora of John Killick et al, recording for which was largely complete by the late 1980s. In contrast, the New Atlas period saw less intensive recording than for the flora and the current atlas project. The difference between native and alien plants also obscures the picture, as many aliens are casual or sporadically recorded. This is well illustrated by the hectad with the city of Oxford (SP50), which has historic records of very many alien taxa from the city's old tips and waste ground. I will let the boffins at BSBI and CEH ponder how to deal with all these complications!

Drilling down to a scale more relevant to local recording, the tetrad coverage was good, indicated by the distribution, colour and size of the tetrad squares in the map shown. There were some solidly recorded areas, especially in the centre of the county, with effort thinning out to the south-east and north-west. Tim Harrison was a great help in improving several hectads in the south-east. The good results for the centre of the county doubtless reflect the very rich and varied sites around Oxford, but some recorder bias is clear, these sites also being much more accessible than in more remote areas (many are nature reserves) and this is where most of the botanists live. This includes myself, and Islip where I live (SP51H) stands out as among the most species-rich tetrads in SP51 together with the unimproved part of Otmoor (SP51R). In fact, Islip is fairly unexceptional compared with most of the nearby parishes (although there are two rivers) and many of the species I have recorded are garden escapes.

Summary of Atlas 2020 results in VC23, by hectads (top) and tetrads (bottom). The colour of grid squares is the number of species recorded (as shown in the legends), and the proportion of the coloured area is the proportion of species re-recorded from the comparison period (i.e. 1987-1999 or pre-2000). The tetrad map compares atlas records to all pre-2000 records.

After having worked so hard on Atlas, I am sure like me you are ready and rearing to start the next project. The BSBI has not yet announced the projects the society will work on over the coming years. However, it is likely that it will be something more ecology or conservation focused than Atlas 2020, for instance habitat surveys or developing site lists and registers. Myself, I'd like to direct some focus on data deficient taxa, but will be continuing the tetrad recording. The meetings I will be running this year will combine elements of these and less intense training sessions. I've already put dates in the calendar for this season — if you'd like to join me and are not already on my mailing list then do get in touch.