How does nest substrate choice impact larval bee health?

Methods

Summary

Research questions and methods:

Question 1:

Does sandstone provide superior insulation, in terms of temperature and humidity regulation, compared to other local substrates?

Field methods:

The first question will be addressed by recording hourly temperature and humidity data in multiple nesting materials. The depths at which measurements are taken will be based on published or unpublished nesting biology records of the bee genus Anthophora. The substrates used will be 1) sandstone banks, both where exposed and shielded from morning sun; 2) non-sandstone banks (likely silt or possibly sand, whichever one is more prevalent across the chosen sites); and 3) the ground (very likely sand). Two depths will be used in the ground, as there is higher variation in nest depth for ground nesters. Ambient temperature will also be recorded in order to compare the trends observed within the substrates to the actual outside conditions.

Question 2:

Do any observed differences in insulative ability of nesting materials actually matter for these bees?

Lab methods:

The second question will be addressed by using the field data to generate environmental profiles to be used on bee nest cells that are separated into incubators. In this way, each treatment will represent a different substrate: 1) sandstone banks exposed to morning sun, 2) sandstone banks not exposed to morning sun, 3) non-sandstone banks, 4) shallow ground, and 5) deep ground. To determine a minimal, ideal level of mortality for this population, a sixth treatment will be kept in ideal lab rearing temperatures for managed bees.

Challenges

1. How will you find enough nest cells for your study? Won’t taking all these nest cells harm the bees’ populations?

I have found over 50 nest sites during recent fieldwork. I will take cells from the largest known nest site (>10,000 nests) to ensure sufficient replicates and to avoid harming population health. Further, the species Anthophora peritomae may be substituted for Anthophora pueblo if insufficient cells are found of the latter, as both excavate and nest in sandstone.

2. Where are you going to find five programmable incubators that can be used for a full year?  

This is not a reasonable goal for a student (or for many professors!). However, the most extreme summer and winter periods are the most likely to tax a developing bee’s metabolism. As such, bees will be exposed to the different treatment regimens for a period of 2-4 weeks, the timing of which will correspond to when they would actually face these challenges during their development. Outside of the treatment periods, the nest cells will all be subject to an environmental regime suitable for general bee rearing. 

3. I read your paper! Anthophora pueblo can delay its emergence! How can you be sure those that don’t emerge aren’t just waiting to emerge in a subsequent year?

To weigh the unemerged larvae, I’ll have to remove them from their nest cells. I’ll be able to check whether or not they’re alive when I open the nest cells, and then I’ll measure their mass either way (since I can then separate out the truly dead and delayed larvae in our later analyses). It will be interesting to see whether or not there are differences in the proportion of bees delaying emergence based upon the different treatments. 

Pre Analysis Plan

Hypothesis 1:

Compared to alternative substrates, I predict that sandstone will heat up slower when temperatures are higher (during the day) and lose heat slower when cooling (at night), thereby reducing the overall temperature changes that nest cells face. For the sandstone bank without morning sun exposure, I expect the substrate to warm more slowly early in the day and cool more slowly at night because of the position of the sun. 

Hypothesis 2:

Humidity is a more complex variable. For ground versus banks, I expect ground substrates to have higher humidity, as rain flows over and settles in the ground (and snow rests upon it). Between sand and silt, I expect more humidity in sand because the particle size is larger and there is more interstitial space between sand particles than there is between silt particles. As sandstone is likely intermediate in interstitial space, because cements (often carbonates) fill some of the space between sand particles, I expect it to hold humidity intermediately between sand and silt.

Testing hypotheses 1 & 2:

The analysis of the comparative insulative abilities of different substrates is relatively straightforward. I will compare the hourly trends of temperature and humidity change between the different substrates and the outside environment. This will reveal both the most extreme conditions encountered in each substrate and the duration of these challenging periods (given the hourly readings).

Hypothesis 3:

I predict that there will be no significant differences in mortality between the substrate-based environmental regime treatment groups, though they will likely be higher than seen in the ideal treatment. However, I do expect significant differences in the masses of the bees from the treatment groups. I expect the following outcome, from highest to lowest average end mass: sandstone bank>ground (deep)>non-sandstone bank>sandstone bank (no morning sun)>ground (shallow). If I am incorrect about mortality effects, I expect mortality to inversely follow that trend from lowest (sandstone) to highest (shallow ground).

Testing hypothesis 3: Following the lab incubator trials, bee emergence will be recorded and compared to the number of unemerged cells to determine mortality levels. To investigate sub-lethal effects, I will compare the masses by treatment for 1) emerged adult bees, 2) adult bees that did not successfully emerge, and 3) unemerged, dead larvae (removed from nest cells), and 4) unemerged, live larvae that are delaying their emergence (removed from nest cells). Future studies will be necessary to decouple the mortality and sub-lethal effects of temperature and humidity for this bee, though I think the nest cells will more easily regulate humidity than temperature.