Fertilisation of grasslands significantly reduces their ability to cope with changing conditions even when they contain a diverse mix of plants, finds a global study involving Oxford University.The research, published this week in Nature, showed that diverse grasslands were more stable over time but that this effect was weakened when the plots were artificially fertilised."More diverse areas are generally more stable because different plants will benefit from different things at different times," said Professor Andrew Hector of Oxford University's Department of Plant Sciences, senior author of the study. "To take a simple example, if you have plants that thrive in heavy rain and plants that prefer lighter rain in the same field, then the total amount of plant matter in the field in a given year will be less dependent on rainfall as both types will balance each other out."The study used an international network of natural grassland sites called the Nutrient Network (NutNet), maintained and monitored by plant scientists across the globe and led by Dr Eric Seabloom and Dr Elizabeth Borer of the University of Minnesota (UMN). At each site, plants were collected, sorted, dried and weighed to count the number of species and total mass of plants each year. These were compared over three years to give values for diversity and stability. Stability was defined as the average annual mass of plant matter divided by the variation in mass between years. The study further tested the effect of fertilizers on stability.Fertilisers are known to drive down grassland diversity over time, but this study found that fertilised plots were less stable even before their diversity decreased. This effect was not expected based on previous results.'"Finding the mechanisms by which the most diverse grasslands are less stable immediately after fertilisation took us completely by surprise,' said Dr Yann Hautier of UMN, lead author of the study. "We certainly expected the grasslands to be less stable after fertilisation, but not in the way that it happened."As fertilisation boosts total yields, it was expected to increase short-term stability by making small variations less significant. However, the study found that fertilisation disproportionately increased yearly variations in yield, particularly in the most diverse grasslands.Dr Hautier will be joining Professor Hector at Oxford University this summer to further investigate this phenomenon at sites in the UK."We now know that fertilisation makes grasslands less resilient to change, even before we see decreases in diversity," said Dr Hector. "This is something that we did not see in artificial grasslands in another project called BioDepth. This is similar to NutNet but uses controlled environments rather than natural outdoor spaces. Now we want to investigate how this effect develops in the longer term and would like to take a look at data from the Rothamsted Park Grass Experiment, the longest-running ecological experiment in the world."The experiment was set up by John Laws and Joseph Gilbert in 1856 to demonstrate the benefits of artificial fertilisers, and has been running ever since. Ironically, it has since told us a lot about the negative impacts of artificial fertilisers on biodiversity. We would like to compare this data to what we saw in NutNet to see how the effects pan out in the long term."