Constraints to carrot seed production in New Zealand : a thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy (PhD) in Horticulture, School of Agriculture and Environment, College of Sciences, Massey University, Palmerston North, New Zealand

Abstract

New Zealand is one of the largest producers of carrot seeds globally. However, carrot seed production is increasingly vulnerable to the effects of climate change, which may force the geographical shifts in cultivation locations. These shifts increase the possibility of genetic contamination from wild carrots, which could have a consequential impact on the cultivar purity of carrot seeds. There are limited studies that have explored the impact of climatic factors and wild carrots on carrot seed production within the context of New Zealand. This literature mainly focused on the effect of climate change on floral characteristics of cultivated carrots and management of wild carrots rather than risk of seed contamination or climate interactions specific to wild and cultivated carrots. The overall aim of this thesis is to investigate two key constraints on carrot seed production in New Zealand, especially those caused by climate change and the existence of wild carrots.

A panel data modelling study was conducted to investigate the effects of maximum and minimum temperature, and precipitation during critical phenological stages on carrot seed yield. This modelling study was carried out by using cross-sections from 28 locations within the Canterbury and Hawke's Bay regions of New Zealand that cultivate carrot seed crops and time series from 2005 to 2022. Findings show that there were significant (p < 0.01) variabilities in temperature and precipitation across different growing phases, except for precipitation at the vernalization phase. According to the analysis of marginal effects, the highest significant effect of minimum (187.724 kg/ha of seed yield reduction for each 1 ˚ C increment) and maximum temperature (1 ˚C rise increases seed yield by 132.728 kg/ha), and precipitation (1 mm increment of precipitation declines the seed yield by 1.745 kg/ha) on carrot seed yield were reported at vernalization, and flowering and seed development, respectively. Results from the modelling study conclusively show that carrot seed production will be negatively affected by climatic change. Meanwhile, the germination phase, including seed germination and seedling emergence, is extremely sensitive to adverse temperature, particularly temperatures above the optimum, conditions and represents a critical stage for the successful establishment of seed crops. However, further studies were needed to accurately determine the impact of temperature on the germination of carrot seeds, including wild and cultivated carrots.

Therefore, germination experiments were carried out to investigate how extreme temperature can affect the seed germination of carrot male line and female line, and wild carrots by using a temperature gradient plate. The findings indicated that the germination percentage, mean germination time, and time required for 50 % germination of wild and cultivated carrots can be significantly (p < 0.05) affected by the interaction effects of temperature and genotypes. The highest (87 %) and rapid germination (6.8 days) of wild carrots was observed at 21 ˚C and 25 ˚C, respectively. The highest percentage of germination was recorded at temperatures of 20 ˚C for cultivated carrots, including male (98 %) and female (97 %) lines; however, these differences were not statistically (p > 0.05) significant. The exponential model was chosen based on statistical criteria to determine the base, optimal, and ceiling temperatures for the seed germination of wild and cultivated carrots. The values of base, optimum, and ceiling temperatures of wild, and cultivated carrots ranged from -0.22 to 2.98 ˚C, 22.11 to 25.55 ˚C, and 37.88 to 38.64 ˚C, respectively. According to the climate change projections, the germination rate (GR50) of wild and cultivated carrots is predicted to increase along with the increasing temperature. Given the comparatively increasing germination rate of wild carrots under projected temperature increments, it is important to manage their vegetative growth and development efficiently. However, further investigations are required to precisely determine the most effective timing and growth stage for managing wild carrots.

Consequently, a glasshouse experiment was conducted to compare and model the vegetative growth pattern of different morphological traits, such as plant height, leaf number, root diameter and length, and shoot and root fresh and dry weight, in both wild and cultivated carrots, to understand appropriate weed management strategies with respect to their growth stages. Studying the vegetative growth pattern of wild carrots helps determine the most appropriate growth stages for implementing effective weed control methods, such as timing of herbicide application or mechanical control. This study comprised two main factors; 1. Genotype (T1: cultivated and T2: wild carrots) and 2. Length of juvenile phase, defined as the time duration between sowing and initiation of vernalization (J1: 12-week, J2: 8-week, and J3: 4-week). The morphological traits studied included plant height, leaf number, shoot fresh and dry weight, root fresh and dry weight, root diameter and root length. The recorded data were analyzed using analysis of variance (ANOVA), correlation, and regression analysis, and principal component analysis (PCA). The interaction effect between ‘genotype × juvenile stage’ has shown a significant (p < 0.05) effect on all the traits except plant height. Shoot and root growth of wild carrots exhibited rapid growth after reaching 8-week juvenile stage (9-11 leaves stage). Results from correlation and regression analysis, especially power regression, indicated that the above-ground morphological traits can be used to determine the characteristics of below-ground parts of both wild and cultivated carrots. PCA presented that morphological characteristics, except plant height, can be used to differentiate wild and cultivated carrots. According to the findings of this study, it is clear that wild carrots grow more rapidly than cultivated carrots. Therefore, it is important to manage the wild carrots at their early growth stages, especially prior to the flowering. Meanwhile, further studies are required to compare and understand the reproductive phase of wild and cultivated carrot genotypes found in New Zealand.

Therefore, an experiment was executed to evaluate the effect of different juvenile phases (J1-12 weeks, J2- 8 weeks, and J3- 4 weeks), and vernalization phases (V1- 12 weeks, V2- 4 weeks, and V3- no vernalization) on floral characteristics and flowering behaviour of cultivated (G1) and wild (G2) carrots, to understand the life history strategies of both wild and cultivated carrots. The findings indicated that the interaction effect between G × J × V on the percentage of flowering and time required for flowering was significant (p < 0.05). Cultivated carrots flowered only under treatments with 12-week vernalization, whereas wild carrots exhibited 100 % flowering across all treatments. Overwintering survival was comparatively higher for wild carrots (94.9 % - 100 %) than cultivated carrots (66.1 % - 98.3 %), likely due to their higher cold stress tolerance and deeper root systems. This study also shows that wild carrots can establish themselves as either summer annual or winter annual in New Zealand, whereas commercial carrots may only be cultivated as biennials for seed production.

The overall findings of this doctoral research demonstrate that climate change can have a significant impact on carrot seed production in New Zealand. As a result, shifting the production regions to wild carrot-prone locations can detrimentally affect the genetic purity of the cultivated carrots since wild carrots can establish as a winter annual and summer annual, and have a strong capability of surviving over the winter. Therefore, it is important to control the wild carrot prior to the flowering, especially at their early growth stages. Further research is recommended to investigate the contribution of pollinators on pollen flow from wild to cultivated carrots and to determine the optimum isolation distance by incorporating the factors like range of pollinators, climatic factors, and density of the wild carrot population.