Discussion

[coreytcallaghan]

This follows on from #39, but can probably start writing some of this.

[wcornwell]

Probably should structure this differently than the results, maybe by the hypotheses from the literature?

[wcornwell]

this is good if you haven't seen it already, i think: https://dynamicecology.wordpress.com/2016/02/24/the-5-pivotal-paragraphs-in-a-paper/

[coreytcallaghan]

Yeah. I reckon there are two general ways to structure the paper, and I have no idea what is best.

1.) Here are a bunch of hypotheses, we perform the biggest and baddest assessment and present the results, allowing any hypotheses likely scenario. Title would be "Urban birds or what-not..."

2.) Title would be "Niche breadth (or whatever else) drives urbanization in birds", in which case the structure would be a bit different. And we would discuss other hypotheses more in passing...

Which one is higher impact? I honestly don't know! And, there is a continuum in my head between these two approaches.

[coreytcallaghan]

Brief outline:

Summary of results

Despite previous studies, we found a significant phylogenetic signal in our urbanization index.

Niche breadth, as measured by generalism in our analysis was consistently among the top significant models

Granivores and insectivores were consistently selected against and significantly negative

X, Y, Z had a greater effect size in the phylogenetically constrained models compared with non-phylogenetically constrained models

Ultimately, this suggests that urbanization selects against specialist species, granivores, insectivores. We also highlight that some x,y,z are critical when accounting for phylogenetic similarity.

Ecological explanation/comparison and contrasts with existing literature

These patterns are best explained by niche separation theory (or something)

This paper, this paper, and that paper found this, but we found that. Further, our results confirm results of this paper, this paper, and that paper.

Ultimately, there are many varying results in the literature

How to best 'design' future studies based on our results (really broad-sweeping)

Technological advances/big data/need for continuous measure of urbanization

We now have the ability to incorporate many more species in models than previous (paper 1, 2, 3, 4)

Citizen science data offers a potential opportunity to address these questions into the future

Further study directions

Temporal considerations (we didn't account for any temporal considerations, but future studies should)

Scale considerations (Do these patterns hold true at multiple spatial scales?)

Conclusion

We provide the largest study to date using a continuous urbanization metric in order to assess biological and ecological traits which predict a species' adaptability to urban environments. Understanding the specific traits which are associated with urban birds allows us to fully understand how to best diversity cities in terms of bird-life as urbanization continues. For instance, in our case, Australian cities should focus on providing habitats which allow ground-nesting, more insects, xx, yy, zzz.

[wcornwell]

@coreytcallaghan moving this to here: https://github.com/cornwell-lab-unsw/bird_urbanness/wiki/Discussion

[coreytcallaghan]

Cool. I'm learning all about these github possibilities.

[wcornwell]

me too. i have no idea about the projects tab, for example

[coreytcallaghan]

We moved anything necessary from here into a wiki, so I'll close this for now. Just have to polish off #39 and then can work on writing some of this.