For a German version click here.

Dario Schulz, Christian Stetter, Javier Muro, Jonas Spekker, Jan Börner, Anna Cord und Robert Finger*

Grasslands make up a significant portion of Europe’s agricultural landscapes, representing 34% of the total agricultural land. They are ecosystems of great importance for many wild plant and animal species. Grasslands provide several regulating ecosystem services such as storing carbon, regulating the water balance and promoting pollination, which is essential for plant reproduction. They also play a crucial role in forage production and livestock grazing, making them essential for food security. However, it is evident that the way we manage grassland, particularly through practices such as mowing, grazing or fertilization, can have a significant impact on the development of biodiversity and the provision of ecosystem services.

Frequent mowing is a common practice in intensive grassland management, often employed to maximize hay or silage yields. While it may increase land productivity per areal unit, it also leads to a decline in the diversity of plant species in grasslands, which is detrimental to the overall health and resilience of the ecosystem. This raises an important question: how can we manage grassland in a way that balances productivity and plant biodiversity?

Previous studies have suggested a negative relationship between management intensity and plant biodiversity. However, the extent of this relationship is heterogeneous and context dependent. There is thus, a clear need to analyse the causal relationships between farming intensity, biodiversity and yield, to quantify the underlying heterogeneity for various agricultural systems and different regions.

In our article (Schulz et al. 2024), we investigate how and where the frequency of mowing, as important indicator of grassland management intensity, influences the number of plant species. Using a remote sensing-based machine learning model (Muro et al. 2022), we predict the plant species richness on more than 1.3 million agricultural grassland plots between 2017-2020. Combining this with mowing frequency estimations (Schwieder et al. 2022) in a Generalised Random Forest framework, we estimate the effect of changing mowing frequency on said plant species richness. In other words, we estimate what species diversity would be expected under a different, hypothetical mowing frequency. To do this, we identify and compare areas that are very similar in their observed environmental characteristics and differ only in their mowing regime. This approach allowed for a direct comparison under consistent conditions, enabling us to estimate what would happen in a different, unobserved mowing regime – the so-called counterfactual situation.

As we would expect, the results show that increased mowing frequency leads to a significant decrease in species diversity. On average, one additional mowing event leads to 1.6 fewer plant species. Although previous studies have already identified the ecological mechanisms underlying this effect, we show how this effect varies regionally and how we can use this variability to improve the effectiveness of nature conservation policy. Our findings reveal significant regional differences (see Figure 1). For instance, a single avoided mowing event (i.e., reducing the mowing frequency by one), has a markedly smaller positive impact on species richness in intensively utilised regions, such as southern Bavaria or Lower Saxony, compared to less intensively utilised regions, such as Brandenburg. We also examine the contextual factors that underpin these differences and find that both site-specific environmental factors and socio-economic factors play a role at the district level.

We then go one step further and ask the question: What is the potential cost of yield loss for more plant species in a given field? To do this, we use a biophysical grass growth model called LINGRA-N (Wolf 2012) to simulate the dry matter yields of hay under the observed management regime. We then calculate the change in yields in the counterfactual scenario with a different mowing regime. In general, the less frequent the mowing, the lower the dry matter yield. Thus, by dividing the predicted increase in plant species richness by the predicted decrease in hay yield multiplied by the hay price, we determined for each field how much it would cost to maintain one additional plant species. Our findings show that an additional plant species translates to a yield loss worth 126 € per hectare and year on average due to a lower mowing frequency. There is a significant variation in these values across Germany, with the opportunity costs of an additional plant species reaching up to ten times higher in intensively utilised regions compared to less productive regions (Figure 2).

This has important policy implications, which we discuss in our article. Among other things, our results suggest that a uniform policy regulating cutting frequency on all grasslands is probably not the best way forward. Rather, the solution is to identify areas where conservation measures such as compensation payments are most cost-effective, based on ecological and socio-economic conditions. By focusing measures on areas where the trade-off between yield and species richness is less severe, we can reduce the overall cost of biodiversity conservation. Our study demonstrates that spatially concentrating nature conservation measures in areas with the greatest impact and the lowest opportunity costs results in significant efficiency gains. 

In a scenario analysis, we show that an optimally targeted policy that compensates farmers for lost yields has the same effect on biodiversity but is up to 60% cheaper than a policy that is not spatially targeted. These results are crucial for achieving global conservation targets such as the '30 X 30' initiative, which aims to protect 30 per cent of terrestrial ecosystems by 2030, and to support the implementation of the recently adopted EU Nature Restoration Law. It is vital that we better understand and manage these trade-offs. With the large-scale application and discussion of possible future improvements, we present a practical and scalable tool. It serves as a preliminary step to help policy makers and land managers strike a balance between agricultural demands and the need to protect biodiversity and safeguard the provision of ecosystem services.

Article (Open Access): Schulz, D., Stetter, C., Muro, J., Spekker, J., Börner, J., Cord, A. F., & Finger, R. (2024). Trade-offs between grassland plant biodiversity and yields are heterogenous across Germany. Communications Earth & Environment, 5(1), 514. https://doi.org/10.1038/s43247-024-01685-0

*Authors: Dario Schulz (European Forest Institute, Bonn, Germany. Previously University of Bonn, Germany), Christian Stetter (ETH Zürich, Zürich, Switzerland), Javier Muro (Thünen Institute, Braunschweig, Germany), Jonas Spekker (University of Bonn, Bonn, Germany), Jan Börner (University of Bonn, Germany), Anna Cord (University of Bonn, Bonn, Germany) and Robert Finger (ETH Zürich, Zürich, Switzerland).

Contact:  dario.schulz[at]efi.int

References

Muro, J., Linstädter, A., Magdon, P., Wöllauer, S., Männer, F. A., Schwarz, L.-M., Ghazaryan, G., Schultz, J., Malenovský, Z., & Dubovyk, O. (2022). Predicting plant biomass and species richness in temperate grasslands across regions, time, and land management with remote sensing and deep learning. Remote Sensing of Environment, 282(1), Article 1. https://doi.org/10.1016/j.rse.2022.113262

Schwieder, M., Wesemeyer, M., Frantz, D., Pfoch, K., Erasmi, S., Pickert, J., Nendel, C., & Hostert, P. (2022). Mapping grassland mowing events across Germany based on combined Sentinel-2 and Landsat 8 time series. Remote Sensing of Environment, 269, 112795. https://doi.org/10.1016/j.rse.2021.112795