# Nutritional and physiological limitations shape the radiation-use efficiency response to legume proportion in C4 grass–legume mixtures

## TL;DR

Adding legumes to warm-climate grasslands boosts productivity up to 30%, but too many legumes reduce efficiency due to nitrogen and physiological limits.

## Contribution

Empirically identifies optimal legume proportions in C4 grasslands for maximizing radiation-use efficiency and productivity.

## Key findings

- RUE was optimized at 26%-30% legume proportion across studies and years.
- Unfertilized pastures showed a 110% overyield in RUE at 30% legume compared to grass-only.
- Exceeding 30% legume proportion led to a decline in RUE due to physiological limitations.

## Abstract

Legume introduction is effective for boosting primary productivity in C4-grass-dominated subtropical and tropical grasslands by overcoming nitrogen (N) limitation and consequently improving radiation-use efficiency (RUE), a key metric underlying plant production. However, an excessive proportion of C3 legumes may negatively affect RUE, especially in warm climates. We assessed the relationship between aboveground RUE and legume proportion of Paspalum notatum Flügge (C4 grass) mixtures with a tropical legume (Arachis glabrata Benth.; 0%–80%) under different defoliation and N fertilizer treatments in two studies over 3 years in Florida, USA. Linking the field data to a conceptual model, RUE was optimized at 26%–30% legume proportion across studies and years. When pastures were N-fertilized, RUE plateaued at 26% legume (0.60 g MJ−1) and linearly decreased with higher legume proportions. When pastures were unfertilized, RUE showed a quadratic relationship with legume proportion, being maximized at 30% legume (1.10 g MJ−1), overyielding the RUE in only-grass and legume-dominated sites by 110% and 86%, respectively. These responses suggest that RUE is N-limited when legume is below 30% in unfertilized canopies and is physiologically limited when legume is above 30% due to replacement of the C4 grass with a C3 legume. These findings provide a robust rationale to target low-to-moderate legume proportions in tropical grasslands for optimizing production and other ecosystem services. We empirically demonstrated that optimum legume proportion is ∼30% in a C4-grass-based tropical grassland compared with previous observations of ≥40% in C3-grass-based temperate grasslands, relevant insights for the design and maintenance of grassland ecosystems.

Linking field data of a warm-climate C4 grass-legume mixture to a conceptual model, we found that the radiation use efficiency, a proxy of primary productivity, was optimized at 30% legume proportion across different studies and years. When these warm-climate mixture were N-fertilized, radiation use efficiency plateaued at 26% legume and then linearly decreased with higher legume proportions. However, when these pastures were unfertilized, the radiation use efficiency quadratically increased with legume proportion, up to an optimum at 30% legume. Beyond this proportion, it decreases following the same pattern as the N-fertilized pasture. These responses suggest radiation use efficiency, and likely primary productivity, are N-limited when legume is below 30% in unfertilized canopies and are physiologically-limited when legume is above 30% due to replacement of the C4 grass with a C3 legume. These findings suggest that low to moderate legume proportions can maximize production and other ecosystem services in warm-climates.