Which plants are the influencers in plant-pollinator networks?

My PhD looked at two invasive plants that has contrasting effects on the native plant-pollinator network. Since then we advanced quite a lot on understanding why superabundant invasive plants with high reward levels can influence others via its shared pollinators, but other less abundant or rewarding exotics don’t. Today, we have a new synthesis paper (Open Access!) formalizing this ideas for any plant species in the network. We analyze lots and lots of plant-pollinator networks to find some generalities. The catch is that we use an index that calculates the potential for one plant to influence another plant. For example, if two plants share only one pollinator and this one do not visit anything else except this two plants, the influence will be very high. On the other hand, if this pollinator also visit lots of other plants, the influence will be lower (see the paper for details). The nice thing is that we can identify some plant traits that make them “influencers”, like plants offering abundant resources and open flowers. It’s a shame that we couldn’t tell (yet?*) if the influence is positive or negative, but at least we can identify key influential plants within the network.

*It may be a way to test for that and at some point we talked about a follow-up, but who knows…

Invasive plants and plant pollinator networks

We have a new article revisiting the topic of my PhD thesis, but with a twist. Invasive plants effects on network modularity. Back then I already explored a bit the effects of invasive plants on the modular structure of the plant-pollinator community, but I never published any result, among other things because I didn’t understand well what I was doing. That’s why when Matthias asked me to join his paper addressing this question with a bigger dataset I was very happy to give it a second try.

Now (6 years later!) I understand two key things way better. First, that invasive plants have different roles in the network than natives, not because they are not native, but because of its different characteristics (i.e. very abundant and generalized). Second, that it is more interesting to understand how the roles that different species play within the network change, than how the overall network structure change, mainly because very different networks can present very similar structures (i.e. nested and modular). I think we nicely present this two points in the paper. See the figure below, and read the paper if you a curious about knowing more.


Winners and losers of land use intensification

We have a new paper out showing which pollinators are more affected by land use intensification and which can cope with it quite well in New Zealand. There is a clear agreement that we should move beyond general species richness patterns, and understand each species specific response. Probably is not surprising that e.g., the invasive Bombus terrestris is doing great while bees in the native genus Leioproctus are struggling with land use intensification. However, most pollinator studies are still mainly based on richness and abundance metrics (Winfree et al 2011). That’s why we really wanted to see not only compositional changes, but also functional changes. At the end, identifying the traits of the winners and losers was the most interesting part of the paper.

As usually happens to me, I entered this article on the analysis phase, which means I can not tell you how cool is NZ, because I’ve never been there. However, I can tell you that the stats cover a lot of ground (may be too much and we lose a bit of focus?) and try to make a good use of functional diversity metrics (see here the code used to separate FD and richness effects) and species identity sensitivity to land use intensification.

Pollinator contribution to yield quality (and my preprint experience)

I already shared a preprint and post about this paper some time ago, but now is officially peer reviewed and online. You can download the final version here: https://peerj.com/articles/328/

My experience with preprints? The publication process at ThePeerJ was super fast (~ 3 months from submission to publication). In this 3 months 84 people visited the PrePrint and only 52 downloaded the PDF. Nobody commented on it. Taking into account that we are 11 authors (who should account for some of this downloads), you may think that the visibility of the paper didn’t increase much from being out there in advance, but I can prove you wrong. Maybe not much people read it, but I was contacted by one PhD student with a question about the paper. She was working on the topic and preparing the next filed season, so for her, reading it in January, instead of in April was useful. Plus, she found it by google-ing about the topic, proving that preprints are discoverable. So, not always by publishing preprints you reach more people, or get amazing feedback, but at least you can reach the right people, and that’s important enough for me.

One more paper showing pollinators matters

We have a new PrePrint up at the peerJ (note that it is not peer-reviewed yet, but already citable) showing that pollinators increase not only yield, but also the quality of four european crops. While the evidence that pollinators are important for crop production is quite strong now, specially after Klein et al. 2007 review and Garibaldi et al 2013 synthesis, I think our paper still contributes to the field by quantifying the contribution to yield (and quality!) in a experimental way along a landscape gradient. Moreover, I think the introduction and discussion is well crafted and points out some aspects that are difficult to cover in short high impact papers (i.e. like our “Garibaldi” science paper). Which points? You will need to read the paper.

You can see the data were collected in 2005, so it has a long, long story I prefer not to dig in. In any case, it ended up in my table and I experienced the pains (and joys) of working with someone else data. That’s why, after waiting 8 years in a messy excel file, I felt that the data deserved to see the light as fast as possible and I pushed to publish it as a preprint. This is an awesome way to make it public probably ~ 6 months earlier than the final reviewed version. I am also happy to try a new Journal that is doing very nice and innovative things. Taking together this preprint and my F1000Research experience, I really think it makes no sense to hide a paper ready to be read until its final version. This can only slow down science. Read more about preprints here.

PS: Also read Klatt et al 2014 paper on strawberries, which spoiled a bit our findings, but is really good.

The paper that made my day

I just read this paper by Hanoteaux et al. (2012), describing the effects of spatial patterns on the pollination success of competing species. They show by using a model that when abundant plant species with attractive flower resources are aggregated, they tend to compete with other less attractive plants, but when they are more uniformly distributed the less attractive plants survive better. The model makes sense from a the pollinator perspective, and is in line with the Circe principle, suggested by Lander et al.

Why I am so happy about that? Well, during my PhD I did a very cool paper showing “Contrasting effects” of two invasive plants (yes, abundant and with attractive flowers, bear with me) on the local plant community. While one invasive species competes with the natives for pollinators, the other one has a facilitation effect. The main problem is that I couldn’t really explain why. But now I can! The invasive plant competing with natives (Opuntia stricta) is clustered, while the one facilitating (Carpobrotus aff. acinaciformis) is uniformly distributed! I would love to be able to update my discussion to add this 2012 reference in my 2008 paper. Would be like citing the future! Would be even better to have actual data on the spatial pattern of each species and test the models. But anyway, science works. I’ve spent 5 years showing a pattern I couldn’t fully explain. Now I have a hint on why this pattern occurs. We are moving forward!


Biodiversity insurance hypothesis in the real world

This year is being great and we have another great publication in Ecology Letters. We use long-term plant and pollinator data to show that high levels of biodiversity ensure plant pollinator matching over time despite climate change.

The story behind the paper starts 2 years ago (yes, it always take time!) when we did a paper showing that in general, plants and bees are advancing its phenology due to climate change at similar rates. The problem of this general patterns is that we don’t present data on any particular case study to show how this “general pattern” translates to a given system. My idea was doing a small follow-up using apple orchards as a case study. I ran the first analysis and saw that indeed, apple flowering and bee pollinators are advancing at similar rates. Cool, We can now provide a case study that validates the pattern observed! But then I went further and tried to see what happens when the main apple pollinators are analyzed one by one. Here the things got interesting because some bee species DO show a phenological mismatch with apple, but the total synchrony is stable at the community level because the effects of individual species cancel out. When I showed the results to Rachael, she immediately related them to the biodiversity insurance hypothesis, and we start working on validating this idea. That meant looking for more data, including a simulation, and a lot of fun reading the biodiversity ecosystem function literature. Is amazing how much of what we know relating biodiversity and ecosystem functioning is based on experiments in grasslands, so applying those concepts to real world trophic interactions was intellectually very stimulating. I like a lot the final paper and I am looking forward to work more on this topic, hopefully with less complex data.

More on Pollinator declines

We have a new correspondence article about bee declines that tries to walk the fine line between a non-helping pessimistic attitude about pollinator declines and an unrealistic optimism. As I said before, I think is easier to defend a black or white position about the pollinator crisis, but I think is time to discuss the grey areas. So here I go:

We show two straight forward things. First, that recent papers showing 50% of bee extinctions and papers showing moderate 15% declines (that’s our paper!) are not reporting conflicting results. Is just a matter of scale. Local scale extinctions in heavily altered habitats translate into population declining trends at the regional scale. To read it in positive, we are still on time to revert this declining trends, because the species are there!

Second, we show that not all species respond equally to global change threads, for example some species love agricultural areas. Most important, seems that the species that thrive in crop fields, are the ones responsable of increasing crop production, so the best current ecosystem service providers (a.k.a. bees that visit crops) may be not as threatened as other bees. But please do not take that as “we don’t have to worry at all”. This is the “grey area” where we need to be clear that highly intensified agricultural areas (e.g. huge almond fields in California) may still suffer pollinator shortages. Similarly, we are talking here about crop pollination, but there is a growing evidence that all species are important to maintain (and stabilize) ecosystem functioning in natural areas. So the good news are only partial.

Read it, is a very short piece and is Open access. If someone is curious about F1000Research, just two lines to say that we choose it for the flexibility of formats they allow, the speed of publication and because I was very eager to see how post peer review works. So far we had two very positive reviewers (which made the article indexed in less than 24 Hours!), but no more comments. Is also a short piece so maybe there is not much else to comment?

Linkage Rules in Plant-Pollinator Networks

I have a new paper in PLoS One with open text, data and code. I like this paper for several reasons. First, is the fruit of the 20% rule. That is, using 20% of my time to side risky projects. I decided to take a webinar on hierarchical models to estimate occupancy last summer. I am not sure why I did it (well, it was free), but the idea of incorporating detectability processes to study animal occupancy was appealing to me. However, I was pretty sure I will not use this models any time soon and I had quite a lot of other things to do, but I decided to “lose” a week anyway attending to it. I didn’t make the connection to apply this concept to plant-pollinator networks right away, but like a month later I got the “aja” inspiration, and I realized that those models can be applied to networks as well. And this resulted in my first single author paper. Not bad!

The basic idea is that doing field work is very time-consuming, and you can not watch all plants species for an infinite time, hence the chances that you detect all pollinators visiting your target plants are low, specially for rare pollinators. Hence, when trying to model what makes a pollinator decide to visit a plant or another, you can not use this raw data because rare species will appear as specialists, when they are probably not. But I show in the paper how to bypass that limitation. You will have to read it to know how, but it involves the above mentioned modelling approach.

I also like the paper for second reason, and is that I think I made a good point on seeing the network as dynamic process. The models use floral traits to predict visitation. Hence, you can use this models to predict which pollinators will visit an invasive plant entering the network. Moreover, I show that I can predict re-wiring of links after a species is removed from the network. Cascading extinction models are quite popular, but do not incorporate this dynamical effects. I know that Jeff Ollerton et al. (who I hope is reading this) have some experimental data that test this “re-wiring” after a plant extinction, so I would love to see if the models predict their data correctly. I think the next big thing will be to incorporate a more dynamic view to p-p networks.

In any case, If you are field biologist don’t be afraid for the “hierarchical modelling” part, (which is not bayesian, but uses likelihood!) and give it a read, because I think I managed to explain it in quite plain English.

Native bees buffer the negative impact of climate warming on honey bee pollination

We have a new paper in GCB lead by Romina. In this paper we do a very cool thing. We characterize the daily activity period of a bunch of bee species and how this activity is modulated by temperature. We show that while honeybees decrease visitation to watermelon at very high temperatures (literature suggest that the reason is that honeybees need to go for water more often when hot, hence have less time to visit flowers), some native bees concentrate its visits on the warmer hours. I think that understanding behavioural differences among species is neat to answer BEF questions.

wtbeeIn addition, we play a bit with future temperature scenarios to see if (all else being equal) visitation and pollen deposition will change with warmer temperatures. We show that the visitation reduction predicted for honeybees is compensated by an increased visitation rate by native bees (taken altogether). Despite this predictions should be interpreted with care, it adds up to the several lines of evidence suggesting that conserving all species is needed in order to have flexible ecosystems able to cope with environmental change.