The relationship between the diversity of herbaceous plants and the extent and heterogeneity of croplands in noncrop vegetation in an agricultural landscape of south China

Abstract

Although agricultural lands hold a small fraction of the biodiversity of natural ecosystems, their sheer size and proximity to remaining natural habitats make them an important target for biodiversity conservation. Non-crop areas, including remnant natural habitats and field margins, often increase biodiversity in agricultural areas, but the importance of cropland heterogeneity (either compositional, the number of crop types, or configurational, their spatial arrangement) is more controversial. Greater knowledge is also needed about agroecosystems in biodiverse areas in the tropics where there are more, smaller-scale farms. We laid out a network of 73 randomly selected 100 m-radius sites in an agricultural area near Nanning, Guangxi Province. Here rice and sugarcane are the major crops, but we encountered a total of 45 less dominant crops. At each site we measured the diversity of herbaceous plants in 10 1-m2 plots placed in uncultivated areas over two summers (2016–2017). The analysis focused on relating plant diversity, for the whole community and for specific categories of species (e.g., native species, animal-dispersed species), with the (a) extent of agricultural land cover, (b) compositional crop heterogeneity (the number and evenness of crop types), (c) length of field margins, a component of configurational crop heterogeneity, all measured at the 100 m-radius local scale, and (d) another measure of agricultural land cover, measured at the 500 m-radius landscape scale. We found that agricultural land cover at the local scale had a strong negative effect. Contrastingly, its effect at the landscape scale was positive and more moderate. The contrast among scales may be due to the quality of our information varying between scales; the dominance of the local scale may be due to the generally low dispersal abilities of herbaceous plants, as reported in other studies. Field margin had a positive effect for native, and especially animal-dispersed species, and may act as an important component of configurational crop heterogeneity, increasing the connectivity among patches of different land-use types, especially for species that disperse using animal movement. Crop compositional heterogeneity had no detectable effect for the whole community or for any species category. Our results confirm that the protection of non-crop areas and configurational crop heterogeneity (i.e., field margins) is as important for herbaceous plant diversity and composition in this tropical landscape, as it is elsewhere.