No abstract

Abstract. Several properties have been suggested to be characteristic of ecotones, but their prevalence has rarely been tested. We sampled five ecotones to seek evidence on seven generalizations that are commonly made about ecotones: vegetational sharpness, physiognomic change, occurrence of a spatial community mosaic, many exotic species, ecotonal species, spatial mass effect, and species richness higher or lower than either side of the ecotone.The ecotones were in a sequence from scattered mangroves, through salt marsh, rush-marsh, scrub, woodland, to pasture. We developed a method to objectively define, by rapid vegetational change, the position and depth of an ecotone, identifying five ecotones. Their positions were consistent across three sampling schemes and two spatial grain sizes. One ecotone is a switch ecotone, produced by positive feedback between community and environment. Another is anthropogenic, due to clearing for agriculture. Two others are probably environmental in cause, and one may be largely a relict environmental ecotone.Sharp changes in species composition occurred. Three ecotones were associated with a change in plant physiognomy. In two, the ecotone was located just outside a woodland canopy, in the zone influenced by the canopy. Community mosaicity was evident at only one ecotone. There were few strictly ecotonal species; many species occurred more frequently within ecotones than in adjacent vegetation, but there were never significantly more ecotonal species than expected at random. There was little evidence for the spatial mass effect reducing ecotonal sharpness, or leading to higher species richness within ecotones. Species richness was higher than in the adjacent habitat in only one ecotone.It seems that supposedly characteristic ecotone features depend on the particular ecological situation, and the ecology of the species present, rather than being intrinsic properties of ecotones.

The nutritive value of pasture is an important determinant of the performance of grazing livestock. Proximal sensing of in situ pasture is a potential technique for rapid prediction of nutritive value. In this study, multispectral radiometry was used to obtain pasture spectral reflectance during different seasons (autumn, spring and summer) in 2009–2010 from commercial farms throughout New Zealand. The analytical data set (n = 420) was analysed to develop season‐specific and combined models for predicting pasture nutritive‐value parameters. The predicted parameters included crude protein (CP), acid detergent fibre (ADF), neutral detergent fibre (NDF), ash, lignin, lipid, metabolizable energy (ME) and organic matter digestibility (OMD) using a partial least squares regression analysis. The calibration models were tested by internal and external validation. The results suggested that the global models can predict the pasture nutritive value parameters (CP, ADF, NDF, lignin, ME and OMD) with moderate accuracy (0·64 ≤ r2 ≤ 0·70) while ash and lipid are poorly predicted (0·33 ≤ r2 ≤ 0·40). However, the season‐specific models improved the prediction accuracy, in autumn (0·73 ≤ r2 ≤ 0·83) for CP, ADF, NDF and lignin; in spring (0·61 ≤ r2 ≤ 0·78) for CP and ash; in summer (0·77 ≤ r2 ≤ 0·80) for CP and ash, indicating a seasonal impact on spectral response.

Abstract. Relative abundance distributions (RADs) are an important feature of community structure, but little is known of the factors affecting which type of RAD is observed in a particular community. We examined the influences of species richness and of spatial scale on the RAD of plant communities. The effect of species richness was examined by analysing simulated communities generated under the Broken stick model, the Sequential breakage model, and a randomized version of Niche pre‐emption model. In all cases, when there were few species in the community the data only occasionally gave the best fit to the model that was used to generate it. With 40–65 species, a best fit was obtained for the correct model in about 75 % of cases, almost irrespective of the model. Effects of spatial scale were examined in data from four dune slacks and from two semi‐arid grasslands, by analysing biomass values at a range of sample sizes. The model that best fitted the whole sample differed between the four slacks and between the slacks and the semi‐arid grasslands. The change in which model of RAD fitted best, as sample size was reduced, varied between sites and between habitat types. At the smallest sample sizes, the Zipf‐Mandelbrot model often fitted, and in the slack sites the Broken stick also, though neither fitted (in the vegetation examined) at larger spatial scales. It is concluded the RAD is affected by species richness and by spatial scale, in ways that currently do not enable simple prediction. RADs can theoretically give information on the processes such as resource partitioning, immigration and competition that have structured the community, but they are a blunt tool for this purpose.