A brief history of our work on understanding, explaining, and predicting variation in food web strucure.
The first quantitative explanation (i.e. mathematical model) for the number of trophic links one would expect to see in a food web. We also showed that when parameterised with independent data the explanation made very nice predictions of what is observed in real food webs (Beckerman et al 2006.). The explanation is based on the contingency foraging model that predicts which of the food items a consumer should choose to consume out of all available food items. The contingency model does this by finding the food items that would maximise energy intake rate, and works with organismal traits such as energetic content of a food item, rate of space clearance, time to handle a food items, and probability of encountering different food items.
The first model able to predict food web complexity and structure. We added to the model size dependency of the organismal traits (i.e. we added allometries of the foraging relevant traits) (Petchey et al 2008). Doing this created explanations for the patterning of trophic links in food webs that were able to very well explain the observed structure of some real food webs. We termed this model the Allometric diet breadth model (ADBM).
The first predictions of temperature effects on food web complexity (connectance). We added to the model temperature dependence of the organismal traits to the ADBM (Petchey et al 2010). The effect of temperature on the number of links (and connectance) in a food web is predicted to depend on variation in the effects of temperature on the different organismal traits.
83% of trophic links predicted correctly when ignoring taxonomy.. When allowed to ignore taxomony the ADBM explained the structure of empirical food webs almost twice as accurately as the equivalent species-based explanation, with the best performance being 83% of the links predicted correctly in the Broadstone Stream food web. (Nasty paywall on this: Woodward et al 2010) (Here’s a link to the article on SciHub, though you should be aware that it may be illeagal to use this resource).
Clarification of what is so different and special about our approach to explaining and predicting food web structure. Peer reviews of Petchey et al 2008 argued that we should not compare our model to previous ones as it would be akin to comparing apples and oranges–we were happy they view our model as being so different/special. We clarified the differences and their implications (e.g. that comparisons with other types of model is rather misplaced) in this article: Petchey et al 2011.
Adpative rewiring predicted by the ADBM stabilises food webs. A study involving model food webs in which species are made extinct and the remaining ones adjust their diets according to the contingency foraging model in the ADBM. This adjustment, also termed rewiring generally leads to fewer secondary extinctions, though not always, depending on species body sizes, and the criteria used for secondary extinctions. (Thierry et al 2011)
Foraging allometries explain variation in food web structure. An theoretical investigation of if and how allomtries of organismal traits (e.g. handling time, space clearance rate) affect the foodweb allometries: the generality‐mass correlation, the vulnerability‐mass correlation and the trophic height‐mass correlation. The results demonstrate how allometries of individual foraging behaviour are linked across scales of organisation. (Thierry et al 2011)
What next? Its now early 2019 and we are starting some new related projects. Hopefully we will have some more points to add here very soon.