by Harry Watkins
We know what towns and cities need: we need more of the right trees in better places. The case for the closer integration of plants into street-level engineering systems is gathering momentum through industry guidance, award-winning best practice, and research. Similarly, the need to find the ‘right’ trees is well known: there is overwhelming research that argues that we need to diversify our urban forests with species that are able to tolerate the complex and changing stress environments that we see in towns and cities, and industry guidance is improving as we learn more about plant traits.
However, this challenge is complicated by a phenomenon not well addressed by landscape architects and commercial horticulture: not all genotypes within a species are equal, making the task of finding the right plants for the right place more difficult. In 2003, fifty years after Crick and Watson published their groundbreaking research, the DNA molecule experienced a midlife crisis: no longer do scientists believe that DNA is a ‘master molecule’ that acts like a blueprint controlling all aspects of development, instead, a more nuanced understanding prevails, seeing DNA as a regulatory system from which different expressions (phenotypes) can emerge. This progress in research has been complemented by simultaneous revolutions in our understanding of epigenetic processes and indeed the understanding of the environment itself as a source of regulatory information rather than a backdrop, leading to the field of research known as eco-devo, or ecological developmental biology. This understanding of the genome as a dynamic system of co-active signals and feedbacks has exciting philosophical and practical applications for CfDE.
This research project asks whether there are ways in which we can identify populations of trees where there is a greater probability of finding genotypes that confer stress tolerance to the phenotypes within them.
For this research project we have chosen to study Magnolias, a genus that is typically seen by horticulturists and landscape architects as being tender, shallow-rooted and having a high water demand: by no means ideal or even suited to the harsh pressures of urban environments but certainly representing an opportunity for diversifying our urban forests. But Magnolia is a promising genus to study as it has a fairly cosmopolitan distribution, suggesting a wide range of environmental adaptations, and therefore hopefully, a wide range of phenotypes for us to investigate. At the same time, developments in landscape engineering mean that we can design the urban environment in ways that suit trees with higher water demands, opening the potential for a wider range of species and genera than ever before. In this light, the genus Magnolia starts to become a possibility for urban foresters.
To unpick the influence of environmental conditions on phenotype expression, we need to see environmental conditions from the plant’s perspective, and the standard approach to evaluating plant provenance is in terms of latitude and altitude. But this approach has its own limits. This shorthand cannot reveal the differences between the conditions at 1,800m altitude on Wa-Shan where E. H. Wilson collected Magnolia sargentiana in 1908, and the conditions at Pac-hsiing-hsien at 2,200m altitude where the Chinese botanist Qu Guiling collected the same species in 1936. After all, it is not the view from a hillside that makes a difference to the plant, rather it is the factors that limit its growth and ability to reproduce. In this way, it is more useful to evaluate population provenance in terms of the amount of water that is available over the course of the year and the amount of light available for photosynthesis during the growing season.
To investigate where in the urban forest it might be appropriate to specify Magnolia species, we started by consulting the horticultural literature that landscape architects most frequently use. We searched in plant encyclopaedias published by the RHS and Michael Dirr, nursery catalogues such as Barchams, Hillier, and Coblands, and industry guidance such as the National Plant Specification. By recording the descriptions of each Magnolia species within each publication and assigning a value of 1-5 for their respective light and water demands, we were able to plot the species on a chart and get a picture of the received wisdom: most Magnolia species require plenty of water and light and as such, probably are not well-suited to urban environments.
However, this picture has a number of problems. Whilst these resources draw on practical experience, few of the species descriptions are based on systematic or controlled trials, and it is not known whether the Magnolias that we see in the UK are in fact representative of the forms that grow in the wild- how does the Magnolia kobus that we see in our gardens compare to its relatives growing in the sub-tropical regions of Kyushu or those adapted to the harsh winters of the island of Hokkaido?
The first step in assessing the environmental conditions to which populations have adapted was to find accurate records of the species’ natural distribution. Interestingly, much of the go-to literature in the Magnolia monographs was not sufficiently detailed for this purpose: we ruled out references that stated, for example, that M. sprengeri is distributed in Hubei Province at 300-900m, and only accepted records with a verified report of a species growing, accurate to village-level or using coordinates. To do this, we carried out an extensive literature review across a wide range of sources:
- Herbaria at RBGE, RBGK, and Oxford University, and JStor’s Global Plants (which hosts copies of herbarium vouchers that have been digitised) and the Chinese Virtual Herbarium,
- Living collections in arboreta and botanic gardens around the UK and northern Europe,
- Records of expeditions held at RBGK and RBGE and written up in journals such as Magnolia Society International, RCMG, The Plantsman, and the International Dendrology Society Yearbook.
- Peer reviewed articles covering community ecology, genome transcription, medical research, and even trials of drone and LIDAR research which revealed plant associations.
In total, we assembled 247 records of naturally-occurring temperate Magnolias across 22 species.
These Magnolia records were used to derive coordinates, which were then used to locate the records on the World Bank Climate Change Knowledge Portal, which hosts world climate data assembled by the University of East Anglia. Once the locations were verified within the terrain model, mean monthly rainfall and temperatures for each record were noted. The mean monthly rainfalls were then summed to produce an Annual Rainfall measurement, and the mean monthly temperatures were used to derive a Warmth Index rating using the methodology of Yim and Kira. Although it was developed over forty years ago, this methodology remains powerful and relevant, because it can act as a proxy both for the amount of light available for photosynthesis during the growing season, and for the length of the growing season itself. This method allows us to describe a lot of information simply.
To identify how these populations varied between and within species, we then plotted each of these records on charts with axes of Warmth Index and Annual Rainfall, as you can see in the gif below: each blue dot is a population of Magnolias. To get an idea of context for these plots, we also carried out these assessments for cities around the world and marked these with red dots: Jakarta (hot and wet), Vancouver (cool and wet), Cairo (hot and dry) and London, which, in a global context, is relatively cool and dry. To add a degree of nuance, we added Nantes for two reasons: firstly, it is a city which is famous for the thousands of Magnolias which grow in the city’s streets and secondly, because the conditions there are similar to the conditions that some climate change models predict we will see in southern England in 50 years time.
The fundamental message that the literature review shows us is in these charts is that species should be understood in terms of ranges rather than positions: ranges in niches that they occupy and ranges in the stresses that they tolerate. Mean positions such as those seen in the horticultural literature are handy summaries, but they only show a part of the picture, which is that in some cases, such as M. virginiana, there is a substantial variation in these species’ ranges. At the same time, the charts prompt questions about why these ranges vary: for example, M. kobus and M. salicifolia are closely related and occupy fairly similar geographic ranges, so why do they occupy different environmental niches?
We should also remember that the plant-based concept of species remains contentious, and that the boundaries between one species and another are by no means fixed: the process of speciation is fluid and the species distinctions we observe at the moment could in some cases be examples of ongoing allopatric speciation- as such, these charts offer a small window into the discussion amongst botanists of whether the range of hexaploid Magnolias (M. campbellii, M. dawsoniana, M. sargentiana and M. sprengeri) in western Sichuan should all be lumped into one large species.
It is also interesting to note that the niches that some species occupy (such as M. biondii and M. sprengeri) have conditions that are very similar to what we see now in parts of the UK, which in turn prompts the question: if they can happily tolerate and reproduce in these conditions, how much stress would they tolerate in the UK before they slow or stop growing and reproducing? There is a very practical answer to this question, which is to test them and find out: so far we have carried out a review of the leaf traits found in temperate Magnolia species using the methodology set out by Andy Hirons and Henrik Sjoman in their review of Acer species, and we are progressing to systematic stress tolerance trials of Magnolia seedlings this year. We will also be carrying out fieldwork in Japan to investigate the range in trait expressions across environmental gradients and match these against the results of this literature review and the seedling trials- our first expedition will be in April 2018, so we will be back soon with more developments.