Effect of species identity and diversity on biomass production and its stability in cover crop mixtures

Abstract

Thanks to positive interactions between species, growing mixtures of cover crops allows improving the ecosystem services provided by cover crop cultivation. In this study, the influence of species diversity but also of species identity and mixture composition on cover crop biomass production and its stability in diverse growing conditions was studied. Several field experiments (varying soil type, preceding crop, soil tillage, sowing density, nitrogen fertilization and spatial replication) were set up in Switzerland during the period 2013−2016. In these experiments, the performance of cover crop species grown as sole crops was compared to that of multispecific mixtures. Part of these experiments followed a simplex design in which four cover crop species were combined together with different proportions, producing a total of 25 mixtures of varied diversity. The other experiments compared sole crop and mixture biomass production in standard randomised block or split plot experiments. Globally, mixtures tended to produce slightly more biomass than the sole crops, with an average between 2 t/ ha and 3.2 t/ha for sole crops and of about 3.5 t/ha for mixtures. Overyielding as well as transgressive overyielding were observed, in 81% and 37% of the cases on average, respectively. However no effect of the level of species diversity within mixtures could be found. Biomass production of cover crops was highly influenced by their growing conditions and by the identity of the species involved, especially for sole crops and bispecific mixtures. The analyses of the simplex experiments allowed to show that species interactions played an important role in biomass production in 7 out of 15 growing conditions, even for a short growing period of about three months. Most of the cover crop mixtures with the highest biomass production had a rather low diversity, i.e. about two species on average, but the identity of the species involved in these mixtures depended on the growing conditions. Our results do not show a strong diversity effect on the biomass production of cover crop mixtures cultivated for a short growing period, but a stronger effect of species identity and of the growing conditions. Mixtures with low diversity generally outcompete more diverse mixtures, but more diverse mixtures offer an insurance effect given the unpredictability of growing conditions during cover crop cultivation. performance due to positive interactions between species. This improved performance of mixtures, called diversity effect, can lead to higher biomass production (Cardinale et al., 2011). It is referred to as 'overyielding' when the mixture produces more than the average of sole crops (Schmid et al., 2008), and 'transgressive overyielding' when the mixture produces more than the best sole crop (Gravel et al. 2012). Overyielding and transgressive over [Gravel et al., 2012] yielding have been shown in grasslands (

Management of cover crops in temperate climates influences soil organic carbon stocks: a meta‐analysis

“…Smith et al (2014) found that diverse cover crop species assemblages yielded more biomass over component monoculture species in an oat ( Avena sativa L.) rotation in the northeastern United States. Similarly, in a study investigating the environmental advantages of bicultures over monocultures, Kramberger et al (2013) found that bicultures supported greater biomass C and N accumulation, which are both important factors for increasing soil C. However, other studies have found that mixtures are equally, but not more, productive as the most productive monoculture species (Finney et al 2016, Wendling et al 2019). In addition, the length of the cover crop growing window (i.e., planting and termination dates) can influence potential for NPP accumulation and design of cover crop species assemblages (Murrell et al 2017).…”

On productivity, resource capture, and yield stability of cover crop species mixtures

“…Species mixtures are expected to have a higher yield stability than pure stands as shown for food crops and grasslands (Isbell et al, 2015;Raseduzzaman and Jensen, 2017). Only a few studies have recently tested whether mixing species of cover crops reduces the chance of crop failure and increases yield stability (Wortman et al, 2012;Smith et al, 2014;Florence et al, 2019;Wendling et al, 2019). The hypothesis that mixing species reduces the risk of crop failure is based on the assumption that if one species 6 fails to grow, the companion species may compensate by filling in the gaps.…”

“…Differences in species responses to weather or soil conditions may result in compensatory growth responses in species mixtures where one species makes up for underperformance of a companion species (Li et al, 2001). Mixing species with different responses to the variable weather and soil conditions in autumn could therefore result in a reduction of risk of the cover crop failing to accumulate enough biomass and provide ecosystem services (Gfeller et al, 2018;Wendling et al, 2019). Furthermore, it is expected that complementary traits, e.g.…”

“…The benefits of mixing species of cover crops have been assessed in several studies, but yield stability was evaluated in only four studies (Wortman et al, 2012;Smith et al, 2014;Florence et al, 2019;Wendling et al, 2019). The term yield stability usually refers to the variability in crop yield over years and environmental conditions, that is: stability is the absence or limited presence of variability.…”

“…Variance and coefficient of variation (standard deviation divided by the mean) are two standard indicators for characterizing variability in ecology and agronomy (Francis and Kannenberg, 1978;Rao and Willey, 1980;Knapp and van der Heijden, 2018). In previous studies on yield variability in cover crops, authors used the coefficient of variation of yield across replicates (spatial stability) and/or across years (temporal stability) (Wortman et al, 2012;Smith et al, 2014;Florence et al, 2019;Wendling et al, 2019). The interpretation of the coefficient of variation can be challenging because it varies not only with the standard deviation, but also with the mean (Doring and Reckling, 2018).…”