Site and Methods

Site Description

Field experiments were undertaken in two semi-natural shrubland remnants located on the Canterbury Plains: an 18 ha privately owned remnant (Spencer-Bower Remnant, SBR), -43o42’91.13” S, 172o43’53.98” E) and a Department of Conservation 2.3 ha Scientific Reserve at Eyrewell (ESR -43o38’31.56” S, 172o19’46.43”). Both were fenced but surrounded by irrigated dairy pasture since the 1990’s for ESR and 2014 for SBR. These remnants comprised canopies of kānuka (Kunzea serotina de Lange and Toelken, Myrtaceae) with a sparse understory containing native and adventive species of vascular plants, including prickly mingimingi (Leptecophylla juniperina C.M. Weiller, Ericaceae) and a pioneer species Pomaderris amoena Colenso, Rhamnaceae . In excess of 70% of ground cover consisted of bryophytes, mainly H. cupressiformeHedwe. var. cupressiforme . The climate of the region is dry with a prevalence of strong north-westerly föhn winds, warm summers, cool winters and < 650 mm rainfall, leading to low humidity and high evapotranspiration rates (Macara 2016). The predominant soil type at the sites is classified as Lismore, a free draining, shallow, stony silty loam of low fertility (pH 5.1, nitrate 0.21 mg L-1, and ammonium 1.64 mg L-1, Olsen P 8.34 µg g-1).

Soil Moisture and Temperature Sampling

Soil moisture was measured during winter and summer months. Six depths of moss (0 - 1; 1.1 - 2; 2.1 - 4; 4.1 - 6; 6.1 - 8; >8 cm) were idenitifed and sampled within the ESR remnant. Three soil cores (5 cm diameter, 7.5 cm depth) were extracted and bulked under each moss depth prior to analysis. Gravimetric soil moisture content was determined on freshly-collected field-moist soil sieved to 4 mm (Blakemore et al. 1987). Soil temperature was measured beneath three different depths of moss (0, 3, 9 cm; n=4) for 19 days during the winter (June 2016) and summer (December 2016 and January 2017). HOBO® Pro v2 weatherproof data loggers connected with 1.8m cables to two external temperature probes were placed 5 cm beneath the soil surface and leftin situ for 3 days to calibrate prior to data collection. Four data loggers failed during winter and two during summer under the 9 cm moss covering.

Available Soil Nitrogen Analysis

Available nitrogen in the soil was determined from fresh soil samples collected under four depths of moss (0, 3, 6 and 9 cm) during the winter and summer months within ESR (n=8) and under three depths of moss (0, 3 and 8 cm) from SBR (n>13). Gravimetric moisture content was determined and a further 4 g sub-sample of field moist soil was extracted with 2M KCl (potassium chloride) to estimate nitrate (NH3--N) and ammonium (NH4+-N) concentrations using standard methodologies (Blakemore et al. 1987). The extracts were analysed using Flow Injection Analyser (FIA) (Foss FIAstar 5000 triple channel with SoFIA software V1.30).

Glasshouse Study

The effect of the moss layer on establishment and growth of vascular plants was investigated in a glasshouse experiment. Seeds of native broom (Carmichaelia australis, n=5) K. serotina andP. amoena (n=10) were cleaned, inspected for damage under a microscope and placed together in 20 trays, each containing one of five treatments (n=4): Eyrewell soil (to 2 cm depth) and four H. cupressiforme moss layers of 2.5, 5, 7.5 and 10 cm depth on Eyrewell soil. Based on earlier findings, P. amoena seeds were pre-treated in boiled water 12 hours immediately prior to use. Trays were located inside an automatically watered mist tent located in a glasshouse. Seeds were monitored weekly for germination and left to establish for 6 months, apart from K. serotina which was left for 4 months. On completion of the experiment, plants were carefully harvested, washed, and oven-dried (72 hrs, 60oC) and soil moisture content and KCl-extractable nitrate and ammonium were determined.

Field Study

Within each remnant, 6 transects of 80 m were identified starting at the fenceline, adjacent to the irrigated pasture, and towards the opposite edge comprising a road (south-north, centre point approximately 50 m). Measurements were taken at the fenceline and at 10 m intervals. At each sampling point plant composition and percentage cover, for vascular and non-vascular plants, was collected using a 1 m2quadrat. Three soil cores (2.5 cm diameter and 7.5 cm depth) were bulked from each quadrat. Soil samples were analysed for moisture and KCL extractable nitrogen. The remainder of the samples were air dried, ground and sieved to 2mm. Plant available P was analysed following the Olsen P method . The resultant Murphy Riley extractant was read at 880 nm on a Shimadzu UV mini-1240 spectrophotometer. Soil pH was analysed using a suspension of 10 g air-dried soil with 25 mL dionised water left to stabilise for 12 h and analysed using S20 SevenEasyTM pH meter.

Statistical Analyses

All statistical analysis and graphical representation was carried out using Minitab® (V 17.2.1) and SigmaPlot (V 12.3). Pearson’s correlation was used to define the relationship between moss depth and soil moisture and temperature. Data were analysed using ANOVA with post-hoc Tukey HSD tests and two-way ANOVA, and Kruskal-Wallis tests for germination and establishment experiments.