Citation: Scott, R.C. & Resetarits Jr., W.J.. 2022.”Spatially explicit habitat selection: Testing contagion and the ideal free distribution with Culex mosquitoes.”
The American Naturalist
Authors:
Reed C. Scott: reed.c.scott@uvm.edu
William J. Resetarits Jr.: wresetar@olemiss.edu

Responsibility for collecting data and writing code goes to R.C. Scott

Directory:
P1S1_Master.csv = data collected + analyzed for experiment 1.

Consists of the following variables:
Array: randomized block that Pool was in
Pool: ID of pool within array
PoolID: ID of pool within experiment. Uses Array # and Pool #
Density: Predictor Variable. Amount of C. restuans larvae added to pool
Day: Number of days after array was set up that data was collected.
Rafts: Response Variable. Number of c. restuans egg rafts collected.
Date: Date on which data was collected.

RawData.csv = Data for experiment 2
Block: Randomized block. Refers to grouping of pools within experiment for statistical purposes
Locality: Within each block, pools were assigned to groups called localities. Each locality consisted of 3 pools in different combinations of Control
and Comptetition
Pool: Pool Number within the assigned locality (a number from 1 to 3)
PoolID: A unique ID assigned to each pool based on its block, locality, and pool number
PoolType: Refers to whether were larvae free (“Control”) or contained C. restuans larve (“Competition”)
Locality Type: The combinations of pools within a locality determined its type. Localities were assigned the label CCC, CCL, CLL, or CCC, where C = Control
and L = Larvae (or competition, as is used in the PoolType column)
Pool.Locality: A combination of pool type and locality type. Used for contrast purposes. For each record input can be read as “PoolType.LocalityType”.
For example, in the first row pool 11.1 is a “Control” pool in the “CCL” locality and thus has a Pool.Locality value of “Control.CCL”.
Day: = days after intial setup that data was collected.
Rafts = Response Variable. Number of c. restuans egg rafts collected.

DMC.master.csv = Data for experiment 3
Block: Randomized block. Refers to grouping of pools within experiment for statistical purposes
Plot: For each block there were 2 “plots” of the experimental design: one with larvae and one without. Plot here is a number assignment (1 or 2) for denoting
the 2 plots within a block.
Locality: Each plot consisted of 3 localites. Locality here is a numeric assignment (1,2, or 3) denoting the 3 localities within a plot
Pool: Pool Number within the assigned locality (1 or 2)
LL: Leaf litter. Refers to the amount of leaf litter (0.1 or 0.5 kg) added to each pool.
PoolType: Pools were assigned a H (High resource, 0.5 kg. leaf litter) or L (Low resource, 0.1 kg. leaf litter)
LocalType: The combinations of pools within a locality determined its type. Each locality consisted of 2 pools. Localities were assigned the label H, M, or L,
where H = High Resource (Both pools were H), M = Mixed (one pool was H one was L), or L = Low Resource (Both for L)
LarvaePlot: Were Larvae added to high resource pools in the plot this pool was in? Values were either Y(yes) or N(no).
LarvaeLocal: Were Larvae added to pools in the locality this pool was in? Values were either Y(yes) or N(no).
LarvaePool: Were Larvae added to this pool? Values were either Y(yes) or N(no).
Pool.Local.Comp: A combination of pool type, locality type, and plot type. Used for contrast purposes. For each record input can be read as
“PoolType.LocalityType.PlotType”. So for the first row, it was a H pool, in a H Locality, in a Y plot, so can be read as “HHY”.
Prop: Proportion of egg rafts that a pool received out of all egg rafts deposited within that pool’s locality
Sum: Total number of eggs collected from that pool.
Day 1 -Day 7 (Columns O-U). Number of egg rafts collected on that day (day = days after intial setup that data was collected).
P1S1.R = analysis for experiment 1. First read in libraries and data. Then aggregate data and construct a linear mixed effects model. Model is tested using dunnett’s
procedure. Then graph.

ComContagion2021.v2.r = analysis for experiment 2. First read in libraries and data. Then aggregate data. Data is then filtered to show only the first 5 days of data
collection. First analysis is patch level, which consists of a simple T-test between “Control” and “Competition” pools. Then graph results.
Next is analysis to test contagion effects. Data is filtered to include only Control pools, and to exclude the LLL locality. Then construct and LMER
with rafts as response variable and localitytype as predictor with block as a random factor. View results and graph. Repeat this process for
competition (or “Larval”) pools. For locality level analysis, construct and LMER with the log transformed sum of rafts (per locality) as the response
and locality type as the predictor with block as a random factor. View results, with Tukey Test. Then graph. Final graph: a comparison of the the
observed vs. expected number of egg rafts in the CCL and CLL localities.

\
DMCContrasts.R = analysis for experiment 3. After libraries and data are read in, the first step is to adjust the data. the predictor variable Pool.Local.Comp is transformed
to a factor, and the Sum response variable is log transformed. Then use the predictor and response to create a linear mixed effect model with block as a
random variable. Once the lmer is constructed, then establish contrasts using rbing (lines 39-45). Finally, test the model using DMCContrasts to test
multiple comparisons. Additionaly comparision is added on in lines 50-53 as a post-hoc test. After this begin graphing.
Graphs comparing the 8 pool types used the same schematic, only differing in the shading of pool types to display comparisons. At line 174 locality level
comparisons begin. Construct a lmer then do a Tukey’s test. Finally, graph.

Code:
All analysis for this set of experiments used R version 4.1.2 (2021-11-01).
OS: Windows 10 x64 (build 19043)