Site description and local climate
The multi-environment trial (MET) was established in the summer of 2018/19 in three commercial potato farms located in three regions (different ecozones) of the North Island of New Zealand i.e., Ohakune (-39.500, 175.459, 593 meters above sea level (masl), central North Island); Opiki (-40.465, 175.481; 3 masl; southwest North Island); and Hastings (-39.624, 176.738; 17 masl; eastern North Island). The Ohakune site is a conventional-irrigated farm, ex-pasture, orthic allophanic – volcanic ash soil (Landcare Research 2020), with an average temperature of 10.2°C, and average annual rainfall of 1,500 mm. The Opiki site is a conventional-rainfed farm, ex-pasture, humic organic soils (Landcare Research 2020), with an average temperature of 13.2°C, and average annual rainfall of 990 mm. The Hastings site is an organic-irrigated farm, ex-cover crop, mottled fluvial recent soil - alluvial sand silt or gravel soil (Landcare Research 2020), with an average temperature of 14.0°C, and average annual rainfall of 770 mm.
Crop management and trial layout
Five offshore-bred potato cultivars – ‘Agria’ (reference cultivar), ‘Hermes’, ‘Fianna’, ‘Taurus’, and ‘Snowden’; and two New Zealand-bred potato cultivars – ‘Rua’, and ‘Ilam Hardy’ from certified seed suppliers were used in the trial. The experimental setup was a Randomized Complete Block Design with four blocks and cultivars were fully randomized in each block. Pre-sprouted seed potatoes 40–60 mm in diameter were manually planted in ridged rows 12 m long with between and within row spacing of 100 and 30 cm, respectively. The grower’s cultural management practices in each site, including land preparation, nutrient and water management, and pest and disease management, were adopted to simulate local growing practices. In the Hastings trial site, being an organic farm, a nylon mesh cover (0.6 mm pore size) was used as a physical barrier to managing the tomato–potato psyllid or TPP (Bactericera cockerelli), which is the insect vector of Candidatus Liberibacter solanacearum that causes zebra chip in potatoes.
Weather data
The atmospheric temperature was recorded using a temperature logger DS1921G-F5# Thermochron (www.maximintegrated.com) in Opiki, and Harvest Air Temperature Sensor (Harvest Electronics NZ Ltd) in Hastings. The soil temperature and daily rainfall (within a 15 km radius of the trial sites) were obtained from CliFlo – New Zealand’s web-based national climate database. The accumulated growing degree days (GDD) at 4.4°C base temperature was calculated using the CIP (2013) formula.
Agronomic and morpho-physiological data
At 65 days after planting (DAP) the plant height (mm), stem diameter (mm), and total plant leaf area (cm2) were recorded. For the total plant leaf area, 10 fully developed leaf sub-samples per plant were randomly selected and weighed, and the leaf area was measured using LI-3100C Leaf Area Meter (LI-COR Biosciences, USA, www.licor.com). The total plant leaf weight was divided by the weight of the sub-samples and multiplied by the measured leaf area to get the total plant leaf area.
The net photosynthesis (µmol CO2 m2s1), transpiration rate (mmol H2O m2s1), and stomatal conductance (mol H2O m² s¹) were measured using LI-COR portable photosynthesis system IRGA LI-6400 model (www.licor.com). At the same time, the chlorophyll content (SPAD value) was recorded using a portable chlorophyll meter CCM-200plus (www.optisci.com). All measurements were done on the 3rd or 4th fully expanded leaf from the shoot tip between 1000 and 1400 h, with eight samples measured per cultivar.
Yield and tuber quality data
At 130 DAP, manual harvesting was done at the different sites. The total yield (t ha− 1), marketable yield (t ha− 1), number of tubers per plant, number of marketable tubers per plant, DMC (%), harvest index or HI (%), specific gravity or SG, and percentage of tuber physiological disorders were recorded.
Tubers were classified as marketable if within the marketable size standards and free or with minimal defects (physiological disorders) or unmarketable if undersized or had severe to very severe defects. Percent tuber physiological disorders (e.g., malformation, second growth, growth cracks, enlarged lenticels, feathering, and vascular browning) was determined by dividing the number of tubers with physiological disorders over the total number of tubers per plant and multiplying by 100.
Ten randomly selected marketable tubers from each cultivar were used to measure the DMC and SG. The DMC was determined by oven-drying the chopped potato tubers at 70°C for five days. The fresh and dry weights were measured, and the DMC was calculated by dividing the dry weight by the fresh weight and multiplying by 100. The HI was estimated using the ratio of tuber dry weight: total plant dry weight and multiplying by 100 (Mackerron and Heilbronn 1985). On the other hand, the SG was determined by first weighing the tubers using a top-loading balance (weight in air) and then weighing them while suspended in water (weight in water) (Haase 2003). The weight in air was then divided by the difference between weight in air and weight in water.
Statistical analysis
All the statistical analyses were carried out in R statistical package version 3.4.0 (R Core Team 2017). The significant difference was analyzed using Analysis of Variance (ANOVA), and mean values were compared among cultivars by using Tukey's HSD test if the overall test was significant.
Genotype x environment analysis
The additive main effect and multiplicative interaction or AMMI method was used, where an integrated ANOVA and principal component analysis (PCA) were applied to analyze the MET (Crossa et al. 1991; Zobel, Wright, and Gauch 1988). ANOVA was used to determine the main effects of genotypes and environments and the PCA for the residual multiplicative interaction among genotypes and environments. After detecting a significant GxE interaction, the data were graphically analyzed by biplot technique to simultaneously classify genotypes and environments (Hongyu, Garcia-Pena, Ara´ujo, and Dias 2014; Kempton 1984; Zobel et al. 1988). Through this method, cultivars with high productivity and wide adaptability, as well as cultivars with specific adaptability (agronomic zoning), were identified.