More stable productivity of semi natural grasslands than sown pastures in a seasonally dry climate

Abstract

In the Neotropics the predominant pathway to intensify productivity is generally thought to be to convert grasslands to sown pastures, mostly in monoculture. This article examines how above-ground net primary productivity (ANPP) in semi-natural grasslands and sown pastures in Central America respond to rainfall by: (i) assessing the relationships between ANPP and accumulated rainfall and indices of rainfall distribution, (ii) evaluating the variability of ANPP between and within seasons, and (iii) estimating the temporal stability of ANPP. We conducted sequential biomass harvests during 12 periods of 22 days and related those to rainfall. There were significant relationships between ANPP and cumulative rainfall in 22-day periods for both vegetation types and a model including a linear and quadratic term explained 74% of the variation in the data. There was also a significant correlation between ANPP and the number of rainfall events for both vegetation types. Sown pastures had higher ANPP increments per unit rainfall and higher ANPP at the peak of the rainy season than semi-natural grasslands. In contrast, semi-natural grasslands showed higher ANPP early in the dry season. The temporal stability of ANPP was higher in semi-natural grasslands than in the sown pastures in the dry season and over a whole annual cycle. Our results reveal that, contrary to conventional thinking amongst pasture scientists, there appears to be no increase in ANPP arising from replacing semi-natural grasslands with sown pastures under prevailing pasture management practices in seasonally dry climates, while the temporal distribution of ANPP is more even in semi-natural grasslands. Neither sown pastures nor semi-natural grasslands are productive towards the end of the dry season, indicating the potential importance of the widespread practice of retaining tree cover in pastures.

Figure 1. Relationships between above-ground net primary productivity and rainfall.

The relationship between above-ground net primary productivity (ANPP) and cumulative rainfall in 22 day periods (ARP22) for sown pastures and grasslands. Sown pastures: n = 44, adjusted r² = 0.74; grasslands: n = 60, adjusted r² = 0.47. The relationships show significant linear and quadratic trends for pastures, in both cases, and grasslands. There were no differences between treatments in the intercepts and the quadratic coefficients but there were differences in the linear coefficient.

Figure 2. Above-ground net primary productivity through time.

Variation of above-ground net primary productivity (ANPP in g m² 22 day period⁻¹) through time for grasslands and sown pastures. Sown pastures: n = 44; grasslands: n = 60. Black vertical lines separate seasons. The dry season was divided into early and late phases denoted by the grey vertical line.

Figure 3. Above-ground net primary productivity for vegetation types and seasons.

Mean above-ground net primary productivity (ANPP) for the different vegetation types in different seasons ± 1 SE, based on LSD test. Significant differences (p<0.05) between treatments are indicated by different letters.

Figure 4. Temporal stability of Above-ground net primary productivity.

Yearly (July 2007–June 2008) and dry season (December 2007–May 2008) temporal stability of ANPP, estimated in five plots with grasslands and at four plots with sown pastures in a sub-humid region of Central Nicaragua. Bars denote ±1 SE, based on LSD test. Different letters in the same time period indicate significant differences (p<0.05) between grasslands and sown pastures.