Geek alert!

Well, I am officially ready to have this baby. I had one last big thing to do before the baby was born, and after a harrowing weekend of data analysis and writing and revising while my sciatic nerve did its best to keep me from sitting in one place long enough to get out a coherent thought, it is done. Twenty-six pages of the culmination of my phd to date are on their way to the USDA's Sustainable Agriculture Research and Education office, reporting on why their money was well spent on me. I sent it out in today's mail and I'm not going to lie, I am quite pleased with myself. So pleased, in fact, that I am going to share some of the highlights with you. I'm not sure this blog is ready to launch into the world of science on its second post, but this will be the last you hear from me on this topic for quite some time, so ready or not, here it comes.

In graphic form!

This is a map of my field sites, in and around the Salinas Valley (near the Monterey Bay). The orange is agriculture, the black is urban or residential, the yellow is grassland or pasture, and the various greens are different kinds of woodland. The little red dots are the farms I worked on from 2006, which are more or less the same for 2007 and 2008, but I'll spare you the nitty gritty details. You can see that some of the farms are right next to or completely surrounded by natural habitat and others are smack dab in the middle of agriculture, and others are somewhere in between. What I'm trying to find out is whether farms derive some sort of pest control benefit from being closer to natural habitat. The goal is both to achieve a more sustainable (and less chemical) means of pest control and to help assess part of the value natural places offer to humans that we are not currently accounting for in our land-use decisions. I'm studying aphids on organic broccoli farms and their predators-- the starring attraction of which are fly larvae in the family syrphidae. You might call them maggots. I call them adorable.

So what am I finding out about aphids and syrphids across this landscape gradient in the Salinas Valley? (Again, I have data from all three years, and I have about 20 different ways to look at these data but I'm just showing you the highlights so you don't get bored.) The original idea was to look across a continuous landscape gradient and see if some kind of threshold exists, like my advisor found for pollinators-- with at 30% natural habitat in the surrounding 1.5 km, you have enough pollination services from wild bees to meet your pollination needs. Unfortunately, it looks like pest control is a little more complicated than pollination. It's a lot easier to detect a difference if I break the data into categories instead of a continuum: "natural" being greater than 50% natural habitat in the 1.5 km surrounding the farm, "agricultural" being less than 15% natural habitat, and "mid-range" being that part in between. The midrange is an enigma to me. I think that's what's screwing up my analyses looking across the whole continuum. But the natural and agricultural ends of the spectrum behave quite nicely. Syrphids start out in much greater numbers at the more natural farms, and while the agricultural farms seem to catch up by the end of the season, it might be too little too late for them to provide control. Aphids, on the other hand, start out at similar initial densities, but keep increasing toward harvest at the agricultural farms, while they seem to be brought under control at the natural farms. The coolest part is the idea that Danny had of plotting syrphids against aphids, broken up by habitat. Or actually, syrphids against previous week's aphids (because, as predators, we expect them to be responding to their prey). For a given previous week's density of aphids, there are fewer syrphids at the agricultural farms than at the natural farms. Furthermore, the slope of the line in this graph indicates the speed with which syrphids respond to an increase in aphids. As you move to the right along the graph (increasing aphid densities), the syrphids at the natural farms increase more rapidly than the syrphids at the agricultural farms. The mid-range has the steepest curve here, but that wasn't the case in all years. Again, the mid-range befuddles me. But all of this is correlative and doesn't necessarily prove that syrphids are providing better pest control at more natural farms.

Enter: the cage study. Last summer I nearly killed myself putting out 96 cages with 100 aphids a piece (which, for those keeping track at home, comes out to 9600 aphids that I meticulously counted out and placed on plants) to set up a little experiment wherein I could control the starting number of aphids across my landscape gradient and see what the resident predators at each farm were capable of doing to that population. I was originally setting out just to test the most agricultural against the most natural sites, but accidentally (or serendipitously) also tested the effect of small-scale on-farm complexity as well. Some of both the agricultural and natural farms had hedgerows and high crop diversity, others were vast monocultures of broccoli. At the end of the experiment I found that complexity at the landscape scale (the more "natural" farms) did improve pest control, but so did complexity at the local scale (the farms with hedge-rows and more diverse cropping systems). In fact, while the hedgerows didn't make much of a difference at the natural farms, they improved pest control at the agricultural farms almost as much as having natural habitat did! This is great news for farmers, because although they don't really have much control over where they fall in the landscape and how much natural habitat is around their farms, they can certainly plant more hedgerows. The only thing I don't know, and the reason I need to repeat this study next year, is whether this effect is consistent over the whole growing season. I did this experiment at the very end of the summer, when aphid populations had already peaked at most farms and predator populations were therefore at their peak. That lagtime in the syrphid populations seen in the earlier graph could mean that predators don't show up at the agricultural sites to control aphids when they first get there. And aphids don't need very much of a head start to get completely out of control.

So that's it. Actually, there's a lot more (including a piece on how mustard may turn aphids into walking mustard-oil bombs), but we'll save that for another day. I am at a good stopping point and can now put this biology away for a few months while I tend to our own little biological miracle waiting to unfold.