Phenological advance of blossoming over the past century in one of the world’s largest urban forests, Gauteng City-Region, South Africa

https://doi.org/10.1016/j.ufug.2021.127238Get rights and content

Highlights

  • The longest phenological and climate change record for southern Africa.

  • A 2.1 day per decade advance in Jacaranda blossom dates is recorded.

  • Temperature have simultaneously increases by 0.1−0.2 °C per decade for Tmax and 0.2−0.4 °C per decade for Tmin.

  • This phenological advance is most significantly driven by winter climatic conditions, of which June Tmax is the most marked, resulting in an advance of 4.2–5.3d/°C.

  • Significant impacts on the future of this tree, which due to its classification as a category 3 invasive may not be deliberately replanted in South Africa.

Abstract

The Gauteng City-Region in the northern interior of South Africa hosts one of the world’s largest and most densely vegetated urban forests. The tree species, distributed across pavements, parks and suburban gardens, comprise a range of indigenous and alien species. The most aesthetically distinct of these is Jacaranda mimosifolia (Bignoniaceae), a purple-blossoming tree introduced from Brazil in the 1800s to beautify the cities. The distinct appearance during flowering and their abundance in the Gauteng City-Region has resulting in reporting of peak flowering events in local newspapers throughout the past century. This provides a valuable phenological record, particularly in southern Africa where phenology is seldom recorded. Analysing these reports of Jacaranda mimosifolia flowering, an advance of 2.1 days per decade is calculated for the period 1927–2019. This occurs against a backdrop of statistically significant annual and monthly temperature increases of ∼0.1−0.2 °C/decade for Tmax and ∼0.2−0.4 °C/decade for Tmin, and non-uniform change in rainfall. This phenological advance is most significantly related to winter climatic conditions, including Tmax, rainfall and frost occurrence. The strongest phenological driver is June Tmax, at a rate of 4.3–5.3d/°C across the City-Region. This advance reflects the response of the tree to regional climate warming, which poses threats to the species and the urban forest in the long term when thresholds for adaptation are surpassed.

Introduction

The Gauteng City-Region, a metropolis which encompasses the municipalities of Tshwane (Pretoria), Johannesburg and Ekurhuleni in central northern South Africa, is home to one of the world’s largest and most densely vegetated urban forests, which covers 16.1 % of the total terrestrial area of the region (Mubiwa and Annegarn, 2013; Schäffler and Swilling, 2013; Symes et al., 2017). The City of Johannesburg alone has been recorded to have over 10 million trees (Hardy and Nel, 2015), many of which are located in the vast suburban areas in a network of green-belts, public parks, pavements and private gardens (Schäffler and Swilling, 2013). Jacaranda mimosifolia (Bignoniaceae), hereafter referred to as ‘Jacaranda’, native to Brazil, were introduced to Pretoria in 1829 (Henderson, 1990; Coetzee et al., 2015) and later Johannesburg during a ‘tree planting boom’ (Schäffler and Swilling, 2013: 248) as ornamentals lining the streets of the suburbs and the Central Business Districts. Pretoria is colloquially termed the ‘Jacaranda City’, with an estimated 33,630 Jacaranda trees (Stoffberg, 2006). Today they are classified in South Africa as category 3 alien invasive species under the Conservation of Agricultural Resources Act 1983 (CARA, 1983) due to their water consumption in the semi-arid region (Stoffberg, 2006), and deliberate replanting is thus prohibited but existing trees are not deliberately removed (Turton et al., 2006). Despite the invasive status, these trees remain popular amongst residents and visitors. The spring blossoms transform the City-Region into a purple vista, and remain a popular attraction for locals and tourists alike. The trees and their blossoms therefore fall into an unusual category of being very popular and a flagship species of a region due to their aesthetics, but because of conservation laws cannot be replanted. This heightens the concerns among the residents and visitors regarding the health and longevity of the trees and their flowers, as once the existing trees have died, they will not be replaced. The flowering season has for the past century been reported in local newspapers, and more recently documented by the general public through the social media platforms of Instagram, Flickr, Facebook and Twitter (Fitchett and Fani, 2018).

Globally, shifts in the timing of phenological events have been recorded across a range of fauna and flora, terrestrially and in the ocean, through a variety of methods including ground-based observations, satellite imagery, digital repeat photography, greenhouse experiments and phenological modelling (Fitchett et al., 2015; Piao et al., 2019). These phenological shifts have been closely linked to changes in climate (Cleland et al., 2007), with individual species responding to specific combinations of increases in temperature, changes in the timing and amount of rainfall, changes in frost occurrence, humidity and sunshine hours (Neil and Wu, 2006; Tang et al., 2017). As a consequence, phenological shifts are highly species and location specific (Fitchett et al., 2015). This results in mismatches between predators and prey, and increasingly hinders cross pollination efforts (Miller-Rushing et al., 2010). The location and species specificity also require a vast body of research to record and quantify the phenological shifts in the wide range of species and large spatial areas to track the effects of climate change, as interpolation from one species or location to another is seldom accurate (Tang et al., 2017; Piao et al., 2019).

Phenological studies remain notably sparse for the African continent (Fitchett et al., 2015; Adole et al., 2016). The majority of those conducted (70 %) have utilized remote sensed imagery (Adole et al., 2016). Only one long-term deliberate ground-based observational plant phenological dataset spanning multiple decades has been identified in South Africa and studied with relation to climate change (Grab and Craparo, 2011). In the absence of deliberate ground-based observations and records, documentary sources, which may include newspapers, diaries and logbooks, that capture phenological events and record their timing at a daily scale have provided a valuable source of data globally (Aono and Saito, 2010; Pfister, 2018). Newspaper records have been demonstrated to be a valuable source of phenological data in the South African context in the case of the timing of the Sardine Run, the annual migration of sardines north-east along the KwaZulu-Natal coastline during winter, an attraction for tourists and key economic event for fisheries (Fitchett et al., 2019). For phenological events to be recorded relatively regularly, they need to be discrete in their timing, have a marked progression to the phenological stage of interest, and be of public interest, whether for economic reasons or due to the spectacle of the event (Aono and Saito, 2010). A pilot study for the Jacaranda blossoms in the Highveld of South Africa confirmed that this methodology can be applied in data collection for terrestrial species (Fitchett and Fani, 2018), and notably that news and social media reporting of Jacaranda blossoms occurred predominantly during the first observations of peak flowering, with strong alignment between data sources. This study follows from the preliminary work of Fitchett and Fani (2018), exploring newspaper records for Jacaranda flowering in the Gauteng City-Region for the period 1919–2019, yielding a record spanning 1927–2019 which represents the longest phenological record to date for southern Africa. This record is compared to climate records for the region to explore the rates of climate and phenological change, and assess the drivers of the advance in Jacaranda flowering.

Section snippets

Setting

The Gauteng City-Region comprises an urban conglomeration of cities and towns, which represent the economic hotspot of South Africa responsible for a third of the country’s Gross Domestic Product (Mubiwa and Annegarn, 2013; GCRO, 2018a). The region covers 2% of the land mass of South Africa (GCRO, 2018b), roughly spanning latitudes of 25.5–26.5 °S, and longitudes 27.7–28.4 °E. Jacaranda trees are grown across the Gauteng City-Region (Henderson, 1990; Coetzee et al., 2015), with highest

Phenological shifts

The composite set of documentary-based phenological records for Jacaranda flowering in the Gauteng City-Region provides a relatively comprehensive chronology (Fig. 2). Data were relatively sparse for the 1920s, and mid-1960s to 1980, with very few records for the 1980s (Fig. 2). Combined, these data reveal a statistically significant advance in the timing of peak Jacaranda flowering at a mean rate of 0.21d/yr for the study period 1927–2019 (r = 0.41, p = 0.0011, df = 62), approximating an

Documentary evidence for phenology shifts: considering accuracy and the value of data rescue

Newspaper records capturing phenological events would comprise what Vellend et al. (2013) class as ‘unconventional data’, obtained through ‘opportunistic science’. These form crucial datasets in regions with a paucity of ground-based phenological records, which include the African continent as a whole and of relevance to this study includes South Africa (Chambers et al., 2013). While newspaper records have been valuable in reconstructing phenological shifts both in South Africa (Fitchett et

Conclusion

This study presents the longest record of plant phenology and climate change for southern Africa, assessing the advance in the timing of Jacaranda flowering in the Gauteng City-Region over the period 1919−2019. We find an advance in timing of flowering at a rate of 2.1 days per decade over the period 1927–2019, which is most strongly related to Tmax during the austral winter month of June. This induces an advance at a rate of 4.2–5.3d/°C. Weaker, yet significant, relationships exist between

Author contributions

JMF conceptualized the research project. KR and JMF collected the data and analysed the data. JMF wrote the first draft of the paper, which KR edited. JMF handled the revisions

Funding

This work was supported by the South African National Research Foundation Freestanding Honours grant awarded to KR; and the DSI-NRF Centre of Excellence for Palaeoscience operational support funding awarded to JMF.

Declaration of Competing Interest

The authors report no declarations of interest.

Acknowledgements

We thank Heritage Andile Fani for willingly sharing data collected in an earlier published pilot study, and to the two anonymous reviewers for their detailed and insightful comments on previous versions of this manuscript.

References (55)

  • R. Adhikari
    (2000)
  • Y. Aono et al.

    Clarifying springtime temperature reconstructions of the medieval period by gap-filling the cherry blossom phenological data series at Kyoto, Japan

    Int. J. Biometeorol.

    (2010)
  • Y. Aono et al.

    Autumn temperature deduced from historical records of autumn tints phenology of maple tree in Kyoto, Japan

    Clim. Biosph.

    (2014)
  • E. Beaubien et al.

    Spring flowering response to climate change between 1936 and 2006 in Alberta, Canada

    BioScience

    (2011)
  • R.I. Bertin et al.

    Climate change and flowering phenology in Franklin County, Massachusetts1

    J. Torrey Bot. Soc.

    (2017)
  • E.M. Bussière et al.

    Patterns of bird migration phenology in South Africa suggest northern hemisphere climate as the most consistent driver of change

    Glob. Chang. Biol.

    (2015)
  • CARA

    Conservation of Agricultural Resources Act 1983 (Act 43 of 1983)

    (1983)
  • L.E. Chambers et al.

    Phenological changes in the Southern Hemisphere

    PLoS One

    (2013)
  • M.P.A. Coetzee et al.

    Ganoderma species, including new taxa associated with root rot of the iconic Jacaranda mimosifolia in Pretoria, South Africa

    IMA Fungus

    (2015)
  • W.J. Conradie et al.

    Effect of different environmental factors on the performance of Sauvignon blanc grapevines in the Stellenbosch/Durbanville districts of South Africa. I. Geology, soil, climate, phenology and grape composition

    South Afr. J. Enol. Vitic.

    (2002)
  • L.L. Dyson

    Heavy daily-rainfall characteristics over the Gauteng Province

    Water Sa

    (2009)
  • L.L. Dyson et al.

    A baseline climatology of sounding- derived parameters associated with heavy rainfall over Gauteng, South Africa

    Int. J. Climatol.

    (2015)
  • J.M. Fitchett et al.

    Advance in Jacaranda blossom phenology in the South African highveld interior: a preliminary analysis

  • J.M. Fitchett et al.

    Plant phenology and climate change: progress in methodological approaches and application

    Prog. Phys. Geogr.

    (2015)
  • J.M. Fitchett et al.

    Progressive delays in the timing of sardine migration in the southwest Indian Ocean

    S. Afr. J. Sci.

    (2019)
  • Y.H. Fu et al.

    Declining global warming effects on the phenology of spring leaf unfolding

    Nature

    (2015)
  • K. Fuccillo et al.

    Assessing accuracy in citizen science-based plant phenology modelling

    Int. J. Biometeorol.

    (2015)
  • Cited by (8)

    • Urbanization driving changes in plant species and communities – A global view

      2022, Global Ecology and Conservation
      Citation Excerpt :

      Studies at a spatial scale detected advanced and more extended phenological events (Cochard et al., 2019; Davis et al., 2016; Fisogni et al., 2020; Kondratyeva et al., 2020; Massetti et al., 2015; Oliveira et al., 2019; Su et al., 2019; Yakub and Tiffin, 2017). Long-term studies that focused on more prolonged periods of observations (Fitchett and Raik, 2021; Lee, 2017; Lindh et al., 2018; Neil et al., 2010; van Vliet et al., 2014; Xingyuan et al., 2016), as well as those that combined both temporal and spatial analyses also detected earlier phenology events (Comber and Brunsdon, 2015; Jeong et al., 2011; Li et al., 2017; Qiu et al., 2017). Analyzes involving a large time scale associated with data between urban and non-urban populations can be useful for comparing the period in which there were changes in phenology and the events of expansion of urbanization in cities.

    • From flowering to foliage: Accelerometers track tree sway to provide high-resolution insights into tree phenology

      2022, Agricultural and Forest Meteorology
      Citation Excerpt :

      Drone-based sensors also offer high spatial resolution phenological data (Klosterman et al., 2018), but obtaining permissions to fly above populated areas poses a hurdle to high-resolution temporal monitoring. Given that species-level studies of tree phenology are constrained by the need for high spatial and temporal coverage (Yang et al., 2020; Badeck et al., 2004), species-specific phenological studies typically still rely on traditional labor-intensive visual observations in the field (Gray and Ewers 2021; Elmendorf et al., 2016; Wohlfahrt et al., 2019; Iler et al., 2013; Inouye 2008) or historical diaries (Fitchett and Raik 2021; Zheng et al., 2013). Visual observations are critical for gathering information about flowering phenophases as exemplified in the national launch of plant monitoring at National Ecological Observation Network sites (Elmendorf et al., 2016).

    View all citing articles on Scopus
    View full text