
Example of some data from this dataset, for illustrative purposes: Table 1, three contrasting plots
Since being set up in 1856 the Park Grass plots have received different fertiliser, manure and lime treatments, many of which have had marked effects on soil pH, which are reflected in species composition, yield, microbial biomass and various soil factors, particularly soil pH. Atmospheric deposition has also impacted on the vegetation and soil. On some treated plots the soil became highly acidified and mats of undecomposed plant material have built up. This dataset reflects these various effects, consisting of soil pH(in water) data for Park Grass plots, 1876- 2023 and also for the mats on the soil surface since 1967:
These soil pH data are discussed in Poulton et al. (2026). Soil acidification occurs through the loss of base cations (Ca, Mg, K, Na) via crop removal or leaching as bicarbonate, chloride, nitrate, and sulphate, leading to hydrogen and aluminium ion accumulation. Acidifying anions derive from fertilisers, organic matter oxidation, and nitrification of ammonium-based fertilisers. Leaching losses depend on rainfall and drainage. Studies on Park Grass experiment examined these processes and developed the RothLime model to predict lime requirements. Optimum grassland soil pH is around 6.0.
Continuous nitrogen in the form of mixed ammonium salts (until 1916) and ammonium sulphate since, caused strong acidification, with topsoil pH near 4 by 1959 despite regular liming. In contrast, applications of sodium nitrate did not acidify the soil. From 1903 onwards, systematic liming trials tested CaCO3 rates and methods. A new scheme was introduced on most plots in 1965. Most plots were divided into four sub-plots which were to receive chalk (if necessary) to achieve and maintain a soil pH of 7, 6 and 5, together with an un-limed sub-plot. Large, but somewhat irregular, chalk applications resulted in considerable fluctuations in pH in some sub-plots but eventually achieved the target pHs, though the sub-plots which had become very acid this took decades (see figure above for illustration). Since 1991, lime applications have followed a three-year adjustment cycle and fluctuations in pH have been minimal. Declining atmospheric acid deposition has since raised rainfall pH and soil pH slightly, moderating previous acidification trends.
Soil sampling Soil samples from the experiment have been collected, analysed, and archived at irregular intervals since the 19th century. All plots were first sampled in 1876 using a metal box sampler, with additional depths taken to 137 cm. Limited sampling occurred in 1870, 1904, 1906, and 1913. Comprehensive sampling was done in 1923 and 1959, with selected plots resampled in later decades. Since 1984, sub-plots have been sampled regularly. Very acid plots developed a 2-4 cm mat of undecomposed material, usually sampled separately. Typically, 15-20 bulked soil cores (0-23 cm) are taken per sub-plot, avoiding edges, with some deeper soil layers sampled every few years.
Soil pH in boiled, deionised water with a 1:2.5 soil: water suspension, a mean of two readings. Soils from selected early samples were re-analysed in 2000-4. Measured on air-dried soil, sieved < 2mm.
Rothamsted Research
This
dataset is
available under a Creative
Commons
Attribution Licence (4.0).
YOU MUST CITE AS: Poulton, P., Perryman, S., Glendining, M. (2025). Dataset: Park Grass Soil pH 1876-2023 Electronic Rothamsted Archive, Rothamsted Research, Harpenden, UK https://doi.org/10.23637/rpg5-PGsoilpH-1876-2023-01
Please review our How to Credit Datasets guidance for more information.
Rothamsted relies on the integrity of users to ensure that datasets are used appropriately and Rothamsted Research receives suitable acknowledgment as being the originators of these data. Please review the Conditions of Use before downloading.
A FAIR data Excel file data-PG-soil-pH-1876-2023-01.xlsx, contains Park Grass pH soil (in water) data for 1876-2023. Frictionless CSV files are also provided.
A Regular Excel file regular-data-PG-soil-pH-1876-2023-01, that provides the data in a readily readable tabular format, containing separate sheets:
This dataset is from measurements led by Andy Gregory, Andy Macdonald Paul Poulton, and previous scientists, and latterly analysed by the Analytical Chemistry Unit, Rothamsted Research, Harpenden, see list of contributors. The data have been collated into this FAIR dataset by Sarah Perryman and checked by Margaret Glendining.
The Analytical Chemistry Unit follows the Joint Code of Practice (JCoPR) and participates in European Quality Assurance programmes. All performance is strictly monitored using certified external standards alongside in-house standard materials. Standards and check samples are monitored and recorded.
New plot ids. From 2025, an additional plot identification number, "new_plot_id" uses a new naming convention assigned retrospectively to all plots since the start of the experiment in 1856. It is intended to show the provenance of the treatments of the plots using ' \ ' and '.' to indicate significant changes, where ' \ ' indicates a liming change and '.' indicates a plot-split for a fertilizer modification. For example, original plot number id., 7\1a now also has the new plot id., 7.1\L\a, where 7.1 indicates it has been split, for a new fertilizer treatment, and \L\a indicates the changes in liming it has undergone (where L a indicates a limed half pre-1964 and a indicating lime to pH 7 from 1965.
For further information and assistance, please contact the e-RA curators, Sarah Perryman and Margaret Glendining using the e-RA email address: era@rothamsted.ac.uk