The Use of Old Records - Some Worked Examples

By J J Day

Nearly, half a million flora records have now been entered into an electronic database. These break down, approximately, as half for each period : 1965-1986 and 1987-2002. Interrogation of the data can now be rapid and change can be analysed with greater versatility.

A) Measuring Habitat Loss using an Indicator Species

Alchemilla filicaulis subsp.vestita, Lady's Mantle

In Worcestershire, this species is a good indicator of certain unimproved grassland communities principally the NVC type MG5c (Rodwell, 1994). This is a threatened and declining habitat throughout lowland Britain. Worcestershire supports a significant proportion of the national resource.

Division of the records into date classes allows a rapid assessment of habitat loss. This is indicative of a loss of approximately 18% in MG5c grassland between 1970 and 2001. Figure 1 indicates monad distribution: records are divided into two date classes - 1970 -1986 and 1987-2002.

Figure 1 Alchemilla filicaulis subsp.vestita 1970 - 2001
= Records 1970- 1986 only
= Records 1987 - 2002

Changes in Species Distribution

Cochlearia danica, Danish Scurvy Grass

The first record, for Worcestershire was in 1977 by a disused brine pit at Stoke Works. It has not been re-found at that site. The population has not contributed to the current expansion.

The pattern of colonisation :-

The rate of colonisation, for a vascular plant, in the period 1989 - 2002 is dramatic. There are currently records from 346 monads. If the remote populations are linked along the roads, then the known spread of colonisation extends to 427 km. This gives a rate of colonisation of 30.5 km./year. or 3.5 m/hour., over this period. Clearly in a favourable habitat, Cochlearia danica is a mobile species, highly efficient at locating suitable micro-niches over fairly long distances.

The pattern of colonisation, on individual roads, is of interest. It can be viewed in detail by examining records on a year by year basis. A number of phases are apparent :-

1. Initial colonisation - often a single colony on a new road system, remote from any other colonies, which remains stable for two-three years. The longest leap along a road system, which can be reliably extracted from the data is 18km. The leaps of initial colonisation, on A roads, have been recorded as up to 17km. A distance of 10-15km. is frequently observed. This suggests that a single colony can broadcast its seed production over a distance of 15km. on major roads, under current traffic levels and management regimes, (e.g. verge cutting, salt application).
2. Several new colonies appear with a disjunct distribution, often several kms. apart, after two-three years.
3. Rapid infill between colonies, all suitable micro-niches occupied. On the most, favourable road systems, such as A-class Trunk roads, infill progresses very rapidly, often over the course of a single year. Phase 2 may be obscured or missing: this occurred on the A449(T). The phases 2-3 on the M5 are not documented in the records. These phases were missed by botanists. Once the seed bank is fully operative and at saturation levels, all suitable micro-niches are occupied.
4. Spread into urban areas; becoming apparent (2001-02). Favoured niches are wall bases, alongside principal roads.

The individual pattern of colonisation for each road is unique.

Seed dispersal appears to be, mainly, on currents of turbulence. Both wind and water are utilised as dispersal agents. Long distance dispersal is probably wind borne - stages 1 and 2. The slower rate of these phases is due to the degree of chance. The rapid phase 3 may be largely water borne as colonies coalesce over micro- and macro-catchments.

The micro-niche requirements can be determined by the pattern within initial colonisations. This is most apparent in the earliest phases of colonisation before seed saturation is reached.
Soils are important. In the colonisation of the A-road network there is a bias towards light well drained soils, such as the Triassic sandstone area and the gravels of river terrace deposits. Clay and calcareous soils are generally avoided, even occasionally, on dual carriageways. On less favoured soils, colonisation of the road network may stall after initial colonisation, or be very slow, such as the A44 east of Worcester.

Other soil characteristics :-

Slope is important. Initial colonies on new roads, tend to occur on down-slopes or in valleys. This is, presumably, due to down-wash phenomenon - of seed, salt and fine silts. A good place to look for new colonies on roads is around bridges over watercourses. Drip from motorway bridges accounts for some colonies, such as on the canal at Offerton and the B4090 at Hadzor. Changes in carriageway direction, such as bends and roundabouts, are frequent sites for initial colonisation. Turbulence, bare ground, silt and seed accumulation are causal factors.

The micro-topography of the profile, across a carriageway, is very important. Bare ground is often a feature. Flat open areas, particularly if puddling is likely, alongside carriageways facilitate rapid colonisation. Loose hard shoulders, such as on A449(T) Worcester northern link, provide prime sites. Central reservations with bare areas are similarly favoured. The 15cm. zone back from kerb tops can be important. In rural, locations farm gateways with bare, puddled, areas by main roads, are preferential niches.

The distribution map can be used, as a quick indicator, of the zone of greatest urban pressure even though colonisation of strictly urban roads and streets is only beginning to show. In urban settings niches of soil are a scarce resource amid a frequently inhospitable environment. The seed bank must have reached high levels to facilitate urban spread.

Roads have a significant local impact, on the physical and chemical properties of soils. Compounds other than salt may also play a role. Concentrations of heavy metal ions may be important, for instance in reducing competition. Current management is creating a novel suite of soils. The vegetation of the British Isles has responded.

Knowledge of the biology of mobility may prove crucial for the retention of biodiversity. Particularly in a landscape with an increasingly fragmented network of semi-natural habitats. Clearly, habitat corridors can be important, in aiding rapid colonisation.

The three figures show the distribution plotted as monads at the end of 1989, the end of 1995, and up to April 2002.

Cochlearia danica all records up to end of 1989

Cochlearia danica all records up to end of 1995

Cochlearia danica all records up to April 2002

C. SITE CHANGE

Old Mill Pool, Beoley SP057681

This pool was described by Fincher (1966) as one of the finest natural sights in the district. In the mid-1980s it was made available for angling. It is relatively small, about 0.6 ha. It is of county significance for wildlife. What has been the impact of the change in management on its natural history interest? Can the old records illuminate change?

The database holds records from six survey dates with varying degrees of completeness. In order to utilise all the records they were plotted on a matrix against time. Points were joined to give a minimum date range for each species. This allows all records, including those for single species, to be used.

Biodiversity

A total of 60 wetland plant species have been found on the site since 1966. Twenty three, approximately 40% of the flora, are lost. This is inclusive of the natural turnover rate, which in aquatic/wetland communities can be high. It is a function of micro-habitat availability, niche number and loading. The totals for previous dates give a better comparison for assessing change. A figure, for total carrying capacity in any given year, is found, by counting down a column. This, partially, offsets the problem of site recording. No single visit will record all taxa present.

There has been a decline from a mean of 40 species for three surveys in the mid-1980s, to 34 species in 2000. This indicates a decline in wetland species diversity of 15%.

Species Richness

Utilising, the county rarity score as outlined in Day J.J. (2001):-

Applying, a similiar method from a national system, National Pond Survey (1993), the Species Rarity Index is :-

The national ranking has fallen from very high quality in the 1980s to a high quality pond in 2000.

The species richness has declined

Table

Beoley Pool Temporal Distribution of Wetland Species in Phytosociological groupings

(* = non-native)

Community change

The losses and change are not evenly distributed, across all habitats/communities.
There has been little change overall in :-

The aquatic community has altered significantly.

Biodiversity has risen but alien species are now dominant.

The quality has declined.

The mean Ellenberg score (Hill, 1999), for nitrogen requirement, has increased from 6.2 in the mid 1980s to 6.4 in 2000. The mean of the lost species is 6. This shift is indicative of an increase in eutrophication, particularily, in phosphate levels. This is a characteristic of angled sites compounded here by the small size of the pond. The likely origins of eutrophication, associated with the fishing include

Alien species are another characteristic problem associated with angling

There is a cycle of inappropriate management. The ecological dynamics of the pond, have not been understood. Rather than working with the capabilities of the resource, the anglers have relentlessly pursued their own vision.
It serves as an example.
This was a yellow water lily pond. These are rare in Worcestershire. Anglers do not like fishing around lily beds. It gives the fish an advantage. They attempted to destroy the water-lilies. Herbicides and clearance were used. The chemistry of the water was altered. Fringed water-lily was introduced. This filled the floating species niche. This became rampant. Herbicides were used. This further disrupted the ecology.
Yellow water lilies are protected by a powerful rhizome, these can grow to significant dimensions, in effect they are subterranean, underwater trunks. They are still present, although much reduced. Meanwhile, the angler's cycle of management continues.

Open swamp species and water margin vegetation - there has been a highly significant loss. These communities have been severely disrupted or entirely destroyed.
The decline of open swamp species is 71% and of inundation grassland species 60%.

These species tend to be easily uprooted annuals or short lived perennials. The loss due, principally, to trampling of edges, disturbance and intentional clearance.

Backmarsh, Fen, Damp Grassland community
The loss is significant, with a decline of 35%, mainly amongst damp grassland species.
These communities are scarce in Worcestershire.

The cause of loss is concentrated people pressure, such as trampling, disturbance, and direct clearance. There is even an annual burn, by a well meaning ornithologist/angler. The problem is exasperated, by the density of fishing pegs and through access onto an island. The later holds an alder fen woodland type W5b (Rodwell, 1991). This is a rare type in the county.
The more robust species have survived best. Small perennials are at greatest risk

Conclusion -

References