Making Forecasts

Jasper Slingsby

What are ecological forecasts?


“Where are we going with all this?”

What are ecological forecasts?

Examples from the Ecological Forecasting Initiative.

What are ecological forecasts?

Can focus on any of:

  • species population sizes
  • plant or ecosystem phenology
  • land cover change
  • tree/forest die-off
  • carbon stocks and sequestration rates
  • water quality
  • disease spread/risk
  • fishing quotas
  • risk of bycatch of species of concern
  • risk of shipping collisions with whales
  • risk of harmful algal blooms
  • etc

Forecasting needs in the Fynbos Biome?

Forecasting needs in the Fynbos Biome?

from Slingsby et al. 2023

Proteaceae as model organisms

Proteaceae as model organisms

Temporal dynamics of publications on South African Proteaceae based on a Web of Science search on 13 June 2012. Figure from Schurr et al. (2012).

Proteaceae as model organisms


Best studied plant family in the Fynbos Biome, with extensive locality and demographic data collected by:

  • conservation authorities (CapeNature and SANParks),
  • citizen scientists (Protea Atlas Project and iNaturalist)
  • researchers since the late 1970s and before…

Proteaceae as model organisms


Used to inform conservation planning, management and monitoring, e.g:

  • protected area planning
  • wildfire management
  • wildflower harvesting
  • climate change vulnerability assessment and monitoring
  • …I’m sure there are others I’ve forgotten

Proteaceae life cycles and demography

The fire-driven life-cycle of Fynbos Proteaceae species, including harvesting, taken from (Treurnicht et al. 2021). Population size/stability are determined by key demographic rates of adult fecundity (size of the canopy seed bank), post-fire seedling recruitment and adult fire survival (blue–grey boxes). These rates are affected in various ways by environmental conditions, life-history traits, density dependence, the timing, intensity and severity of fire, wildflower harvesting, etc.

Proteaceae as management indicators


We use our knowledge of Proteaceae demography for management of Fynbos in two ways:

  1. Species level - for setting guidelines for sustainable wild harvesting of their inflorescences
  2. Ecosystem level - to help determine acceptable fire return intervals

Proteaceae as management indicators

Guidelines are currently set by “rule of thumb”1:

  1. Wildflower harvesting: “[There should be no] harvesting until at least 50% of the population had commenced flowering, a harvest of up to 50% of current season flower heads after this stage, and no harvesting at least one year prior to a prescribed burn” (Wilgen et al. 2016)
  1. Fire return intervals: “No fire should be permitted in fynbos until at least 50% of the population of the slowest-maturing species in an area have flowered for at least three successive seasons (or at least 90% of the individuals of the slowest maturing species in the area have flowered and produced seed). Similarly, a fire is probably not necessary unless a third or more of the plants of these slow-maturing species are senescent (i.e. dying or no longer producing flowers and seed).” (CapeNature, n.d.)

Proteaceae as management indicators

Problems with the rules of thumb?

Variation in demographic rates of 26 serotinous Proteaceae species of seeder and sprouter life-history types across their distribution range (Treurnicht et al. 2016). (a) Adult fire survival; (b) Individual fecundity (F); and (c) Per-capita recruitment rate (R).

Problems with the rules of thumb?

Adult fire survival: Species differ in their reliance on seed…

  • sprouters have high adult persistence through fires and need fewer new recruits from seed
  • seeder adults are killed by fire, so populations depend entirely on recruitment from seed

Problems with the rules of thumb?

Individual fecundity: Species vary in their fecundity

  • total number of seeds per plant
  • = number of inflorescences produced multiplied by the number of seeds per inflorescence

Problems with the rules of thumb?

Per-capita recruitment: species vary in seed viability and seed-specific recruitment success

  • viability depends on pathogens, seed predators, age
  • recruitment depends on seed properties, rainfall etc during establishment, dispersal, etc.

Problems with the rules of thumb?

Intraspecific variation in (a) fecundity and (b) recruitment in response to range-wide variation in fire return interval (time since fire), adult population density and soil moisture stress (% days with soil moisture stress) for Protea punctata (Treurnicht et al. 2016).

Problems with the rules of thumb?


  • intraspecific variation in fecundity and recruitment along climatic, soil, fire regime, population density, pollinator availability and other gradients

  • interspecific variation in this intraspecific variation

    • i.e. species differ in their population-level responses to climatic, soil, pollinator availability and other gradients

Proteaceae as management indicators?



What do we do if the rules aren’t as simple as we hoped?

Demographic models and population viability

Sensitivity of Proteaceae species and populations to different wildflower harvesting scenarios can be assessed with demographic models (population viability analysis) (Treurnicht et al. 2021).

Demographic models and population viability

Intraspecific variation in sensitivity to harvesting depicted as maps for four different species with pink dots highlighting where the change in population-level extinction probability (the difference between extinction probabilities under 0% and 50% harvesting) is greater than 0.1 (Treurnicht et al. 2021). The white and black areas depict species-specific occurrence records and the geographical distribution of all Proteaceae in the Cape Floristic Region, respectively.

Demographic models and population viability

Interspecific variation in sensitivity to harvesting depicted as the proportion of populations per species that are highly vulnerable to harvesting (Treurnicht et al. 2021).

Demographic models and population viability


The model revealed:

  • the current harvesting guidelines (based on rules of thumb) can greatly increase the probability of many populations going extinct!!!
  • some surprises! e.g. even some resprouter species can be highly sensitive to flower harvesting!

What about changes in climate or fire?

Demographically driven predictions of species’ distributions

Projections of the change in population growth rates of Protea repens under different scenarios. (a–b) Reducing (increasing) the observed fire return time by 4 yr. (c) Variation of mean population growth rate as a function of fire return time. The horizontal dashed line indicates where the growth rate is stable (= 1). (d) The difference between present day predictions and projections under future climate change scenario with temperature increased by 1 degree and precipitation reduced by 10%. Figure from Merow et al. (2014).

Near-term iterative ecological forecasts?

We can forecast Proteaceae responses to harvesting, wildfire and changing climatic conditions, but these aren’t set up as near-term iterative ecological forecasts…

The iterative ecological forecasting cycle in the context of the scientific method. From lecture by Michael Dietze.

Near-term iterative ecological forecasts?

  1. They are either not specific about when they are forecasting to
  • We need near-term forecasts, e.g. 5-10 years into the future or the next fire cycle

  1. We need to coordinate data collection and centralize data management so that it can feed data into the modelling workflow.

  2. The workflow must be able to ingest and assimilate new data and produce new forecasts automatically.

  3. The models must characterize and propagate uncertainty so we can

  • indicate the uncertainty to the decision maker
  • focus data collection and model development to reduce the uncertainty in forecasts.

Near-term iterative ecological forecasting



“The need to start forecasting is now; the time for making ecology more predictive is here, and learning by doing is the fastest route to drive the science forward.” - Dietze et al. (2018)

References

CapeNature. n.d. “What a landowner needs to know about fire management Fact Sheet.”
Dietze, Michael C, Andrew Fox, Lindsay M Beck-Johnson, Julio L Betancourt, Mevin B Hooten, Catherine S Jarnevich, Timothy H Keitt, et al. 2018. “Iterative near-term ecological forecasting: Needs, opportunities, and challenges.” Proceedings of the National Academy of Sciences of the United States of America 115 (7): 1424–32. https://doi.org/10.1073/pnas.1710231115.
Merow, Cory, Andrew M Latimer, Adam M Wilson, Sean M McMahon, Anthony G Rebelo, and John A Silander Jr. 2014. “On using integral projection models to generate demographically driven predictions of species’ distributions: development and validation using sparse data.” Ecography 37 (12): 1167–83. https://doi.org/10.1111/ecog.00839.
Schurr, Frank M, Karen J Esler, Jasper A Slingsby, and Nicky Allsopp. 2012. “Fynbos Proteaceae as model organisms for biodiversity research and conservation.” South African Journal of Science 108 (11-12): 12–16.
Treurnicht, Martina, Jörn Pagel, Karen J Esler, Annelise Schutte-Vlok, Henning Nottebrock, Tineke Kraaij, Anthony G Rebelo, and Frank M Schurr. 2016. “Environmental drivers of demographic variation across the global geographical range of 26 plant species.” Edited by Roberto Salguero-Gómez. The Journal of Ecology 104 (2): 331–42. https://doi.org/10.1111/1365-2745.12508.
Treurnicht, Martina, Frank M Schurr, Jasper A Slingsby, Karen J Esler, and Jörn Pagel. 2021. “Range‐wide population viability analyses reveal high sensitivity to wildflower harvesting in extreme environments.” The Journal of Applied Ecology 58 (7): 1399–1410. https://doi.org/10.1111/1365-2664.13882.
Wilgen, Brian W van, Jane Carruthers, Richard M Cowling, Karen J Esler, Aurelia T Forsyth, Mirijam Gaertner, M Timm Hoffman, et al. 2016. “Ecological research and conservation management in the Cape Floristic Region between 1945 and 2015: History, current understanding and future challenges.” Transactions of the Royal Society of South Africa 71 (3): 207–303. https://doi.org/10.1080/0035919X.2016.1225607.