|
Environmental Change
Network
|
ECN SITE LINES: ROTHAMSTEDRothamsted and the ECNRothamsted was founded in 1843. It is the oldest continuously-functioning agricultural research station in the world, and is ideal as an Environmental Change Network terrestrial site. The estate is on the west side of the town of Harpenden in Hertfordshire, 40 km NNW of London (OS map reference TL 134137) on a gently undulating plateau 95-134 m in altitude. The soils are well-drained to moderately well-drained flinty silty loams over clay-with-flints and/or chalk. Rothamsted farm and most of the surrounding area is taken up with arable agriculture, and has been for many years: the Romans were farming here 2000 years ago, and the remains of a Roman temple were uncovered close to the site of the present farm buildings. A number of long-term experiments, known as the ‘Classical Experiments’, were established at Rothamsted in the second half of the 19th century. These cover cereal growth, grassland management and woodland regeneration. Backed up by long runs of meterological, soil chemical and biological, and plant composition data, they are a unique record of environmental change. They also show the great value of carefully controlled, continuously monitored experiments. Broadbalk This is the most famous of the Classical Experiments. It was first sown to winter wheat in 1843, and harvested in 1844. Wheat has been grown on all or part of the field every year since. The experiment was set up to compare different organic manures with inorganic fertilizers. The treatments were varied in the first few years, but since 1852 a more permanent scheme has been established. Some changes have been made over the years to ensure that the experiment remains relevant to modern agricultural and environmental problems, including the use of modern varieties of wheat. Great care is taken to ensure that these changes do not compromise the long-term integrity of the site. The yields of grain and straw have been recorded yearly, and samples taken for chemical analysis. These samples, together with those from the other Classical Experiments, are archived at Rothamsted and are still available for analysis today. The Broadbalk soil is a heavy clay loam overlaying chalk (Batcombe series). The soil has been sampled and archived irregularly since the start of the experiment, but in recent years a scheme to sample and store it on a five-year cycle has begun. The samples have been used to look at changes in soil organic matter, levels of radionuclides pre- and post atomic age, the build-up of various heavy metals, and increases in organic compounds such as Dioxins and PCBs. On Broadbalk, a system of tile drains was laid in 1849 and, although not intended for experimental use, they were opened at the junction with the main drain in 1866 to sample drainage from each plot. The drains have been used over the years to look at losses of nutrients, e.g. nitrate and phosphate in drainage waters. One of the continuous cereal sections on Broadbalk that has inputs of inorganic fertilizer is being used to provide data for the ECN cereal protocol. Park Grass This Classical experiment was started in 1856, but it is thought that the field was in pasture for up to 200 years before that date. It was established to examine the effect of different fertilizers on the production of hay. However, it soon moved beyond this initial objective, and developed into an experiment that shows the effect of different types and amounts of fertilizer and lime on botanical populations. The boundaries of the plots are still clearly defined, with the transition area between each plot being about 60 cm. This has shown that in flat, undisturbed grassland there is little sideways movement of nutrients. The plots are cut for hay in June, and cut again in October for silage. Hay yields were calculated from the whole plots up to 1960, since when the yields have been calculated from two silage strips cut from each plot. The position of these strips is varied each year to prevent any systematic effect on the sward. The rest of the plot is still cut for hay so that the grass seed is returned to the plots. Small amounts of lime were applied to Park Grass twice in the 1870s and 1890s; a regular liming scheme was introduced in 1903, with lime added to half of most plots every four years. This continued to 1964 when the plots were further divided into 4 sub-plots. The a, b, and c sub plots are now limed to a pH of approximately 7, 6 and 5, respectively; the d sub plot is left unlimed. The pH of the soil on Park Grass shows some very interesting trends. The overall pH of the experiment was 5.7-5.8 in 1856. Where fertilizer is added as ammonium sulphate the plots have become very acid, with the pH of the surface soil on some plots being as low as 3.9 (in water) where lime has not been added. The steady acidification has resulted in an excellent example of a Chemical Time Bomb when aluminium, suddenly released from the soil when the pH reaches a critical value of about 4.2, is taken up in large amounts by the hay. The fertilizer and lime treatments have also affected the species composition of the plots. The most acidic plots are dominated by one or two species e.g. Holcus lanatus and Anthoxanthum odoratum, and are similar to an upland grassland. The unfertilized ‘Nil’ plots are still the most diverse, and similar to a lowland pasture, containing 40 to 60 species. Broadbalk and Geescroft Wildernesses A small part of the west end of the Broadbalk Experiment (0.2 ha) was fenced off in 1882, and a larger area of a bean field (1.3 ha), called Geescroft, was fenced of in 1886. Both areas have been left untended, and on both a mixed deciduous woodland has developed. Broadbalk had been heavily limed in the early 19th century (prior to the start of the experiment) by digging chalk from dells excavated in the field. As much as 100 t CaCO3 per acre were spread by barrow and spade on the soil. As a result, Broadbalk Wilderness still has a pH of about 7. Geescroft Field had not previously been limed; it started with a pH of about 7 and now has a pH of about 3.9. This has provided an interesting comparison of the species composition and regeneration on the two Wilderness sites, as well as sites for studying the long-term effects of acid rain on soil acidification, weathering and the mobilization of toxic metals. Geescroft Wilderness is also used for pitfall traps, and contains two sets of nitrogen dioxide (NO2) diffusion tubes. The Insect Survey Rothamsted is fortunate to be in possession of some of the longest runs of biological data in the country, and is highly suited to be a ‘lead site’ in the ECN. Two long-standing entomological interests at Rothamsted came together with the foundation of the insect survey in the early 1960s. One of these was a background of work on insect migration, including aphids, butterflies and moths. The other was expertise in the sampling and quantitative analysis of aerial populations of insects. A light trap, designed by C.B. Williams in 1948, was operated at the edge of Barnfield, a long-term root crop experiment, between 1933 and 1937, and again between 1944 and 1950. All macrolepidoptera, the larger moths, were counted and identified. These data was used to study the effects of weather on insects, and to develop and quantify the concept of diversity based on the observed frequency distribution found in the samples. Suction trapping was also developed as a quantitative method for studying aerial insect populations. The traps were operated at different heights to examine the importance of wind-borne migration in insects, and establish the form of the relationship between insect density and height. The ECN at Rothamsted As there are many different Classical experiments offering a variety of ecological conditions at Rothamsted, it was decided from the start of the ECN monitoring programme that the whole of the farm would be regarded as the Targeted Sample Site (T.S.S.). As the farm covers such a large area, and certain protocols have to be carried out on several different Classical Experiments, the job of site manager is split. Paul Hargreaves in the Soil Science Department, deals with the physical and chemical protocols, and John Bater in the Entomology and Nematology Department covers the biological protocols. The majority of the protocols have focused on the Park Grass experiment, as this is of most botanical and ecological interest. Many of the insect protocols are sited on the western side of Park Grass, with the ECN moth light trap on the northern edge. The soil solution probes are situated on the eastern side of Park Grass in a new, unfertilised ‘Nil’ plot that has been specially set up for the ECN. Six sets of NO2 diffusion tubes are situated on the Park Grass experiment, with three sets on the western side, close to the trees and road, and three on the more open eastern side of the field. The transect for the rabbit protocol follows the western and southern edges of the experiment. The River Ver The River Ver runs through the western side of the farm. It had been dry for several years because of ground water abstraction further up the valley. However, since the ECN started, pumping has ceased and the river has begun to flow again. Regular sampling began in June 1994 and flow data have been obtained from the NRA to back up the chemical analyses. Further work in collaboration with the NRA (soon to become part of the EA) to monitor flow and chemistry is planned. Nitrogen Dioxide Diffusion Tubes The ECN protocol for NO2 tubes calls for one set of tubes to be exposed every fortnight at each site. As the whole of the farm is included in the ECN monitoring scheme, it was felt that a larger number should be set up to look at variability across the farm. Thirteen sets of tubes have therefore been set up, centred on the Classical Experiments. There are two sets in Geescroft Wilderness to look at the effect of woodland on levels of NO2. Six sets have been placed around Park Grass to look at spatial variation on a small scale and the effect of a busy road passing close to the edge of the experiment. Three sets have been placed around Broadbalk, two sets in the hedge close to the pitfall traps and one set close to the Wilderness and the farm. This latter set is to observe any effect from the storage and movement of fertilisers and manure. The other two sets have been placed in the meteorological enclosure. All the tubes have been running for over two years and are providing interesting data on the effects of different vegetation types and climate on NO2 levels. Biological Protocols Prior to the start of the ECN in 1992,data were already being collected for two major national censuses - the Common Bird Census undertaken by the British Trust for Ornithology (BTO), and the Common Butterfly Census organised by the Institute of Terrestrial Ecology based at Monk’s Wood. Both of these protocols have been adopted by the ECN and both are in their fifth year of receiving data from Rothamsted. A team of twelve volunteers carries out the bird census over a twelve-week period. To date, buzzard, wheatear, yellow-wagtail and breeding tree sparrows have been the highlights of the census. A two-mile, twelve-section transect makes up the butterfly census area, with sections passing through varies habitats such as an abandoned narrow-gauge railway line and open, managed farmland. The cropping patterns on the farm are showing quite marked effects on the butterfly population, highlighted in 1992 when nearly 300 large whites were recorded on one section that passed near a rape crop. Numbers dropped dramatically in subsequent seasons when crops changed. In 1995, the first white-letter hairstreak was recorded. Tipulid larvae, spittle-bug nymphs and adults have all been monitored successfully this season, although the hot dry summer made catching adult spittle morphs much more difficult than usual. Carabid beetles and harvestmen have again been caught in large numbers at all three pitfall-trap locations - Broadbalk, Park Grass and Geescroft Wilderness. Fortunately numbers of small shrews and voles caught in pitfall-traps have been greatly reduced by the use of chicken wire cylinders fitted over the traps. Carabid identification is now being checked for our other ECN sites: at Porton Down, two species of uncommon carabid have been recorded in reasonable numbers. It should be possible to start comparing leg colour morphs of Pterostichus madidus (Coleoptera: Carabidae) between sites after the 1996 season. Frog spawn monitoring showed an increase in the number of spawn masses produced this year, although the hot, dry summer almost dried out the pond and topping-up was necessary. Spawn masses were first observed on 11 March 1995, and the final spawn mass laid on 6 April. In total, six adults, two females and four male frogs, were observed this year. Although the incidence of red-leg disease was reported at other local ponds, no incidence was observed at Rothamsted. Small numbers of pipistrelle bats were recorded at Rothamsted Manor, and at the settling reservoir on the edge of the farm. A local bat group is interested in our census because they are hoping to monitor the effect of a local motorway widening scheme on populations of bats near the road. The second part of the rabbit census is nearly complete and numbers remain stable with little or no evidence of myxomatosis
For those interested in reading more about the Classical Experiments and the history of Rothamsted, the following publications are available:
|
