Table 1.
Moderator analysis of intercropping influence on summary effects of grass yield.
Table 2.
Moderator analysis of intercropping influence on summary effects of forage quality (acid detergent fiber, neutral detergent fiber, crude protein, and in vitro dry matter digestibility).
Fig 1.
Weighted, overall summary effect sizes (response ratios) for NPP from legume-intercropping relative to sole grass fertilization rate in kg N ha-1 (a) and relative nitrogen fertilization (b).
Negative values indicate inhibition from symbiosis, positive values indicate positive changes from the interaction. Change refers to raw percent affect in the effect size induced by legume-intercropping. Horizontal bars are 95% confidence intervals of the subgroup (moderator level) summary effect. n is number of studies contributing to the effect size. P value is the probability that the moderator level was statistically not different from zero.
Fig 2.
Weighted, overall summary effect sizes (response ratios) for primary productivity from legume-intercropping on grass photosynthetic pathway (a), legume life cycle (b), legume seasonality (c), and number of legumes in a mix (d).
Negative values indicate inhibition from symbiosis, positive values indicate positive changes from the interaction. Change refers to raw percent affect in the effect size induced by legume-intercropping. Horizontal bars are 95% confidence intervals of the subgroup (moderator level) summary effect. n is number of studies contributing to the effect size. P value is the probability that the moderator level was statistically not different from zero.
Fig 3.
Weighted, overall summary effect sizes (response ratios) for primary productivity from legume-intercropping based on grass genera (negative values indicate inhibition from symbiosis, positive values indicate positive changes from the interaction).
Change refers to raw percent affect in the effect size induced by legume-intercropping. Horizontal bars are 95% confidence intervals of the subgroup (moderator level) summary effect. n is number of studies contributing to the effect size. P value is the probability that the moderator level was statistically not different from zero.
Fig 4.
Weighted, overall summary effect sizes (response ratios) for primary productivity from legume-intercropping per Koppen climate classification (a), soil texture (b), and establishment period since 1971–2016 (c) (negative values indicate yield inhibition from symbiosis, positive values indicate positive changes from the interaction).
Change refers to raw percent affect in the effect size induced by legume-intercropping. Horizontal bars are 95% confidence intervals of the subgroup (moderator level) summary effect. n is number of studies contributing to the effect size. P value is the probability that the moderator level was statistically not different from zero.
Fig 5.
Weighted, overall summary effect sizes (response ratios) for crude protein from legume-intercropping based on system management including row spacing (a), photosynthetic pathway (b), legume life cycle (c), legume seasonality (d), and soil texture (e).
Negative values indicate inhibition from symbiosis, positive values indicate positive changes from the interaction. Change refers to raw percent affect in the effect size induced by legume-intercropping. Horizontal bars are 95% confidence intervals of the subgroup (moderator level) summary effect. n is number of studies contributing to the effect size. P value is the probability that the moderator level was statistically not different from zero.