Methods

Lab 2:

Methods & Data Collection

Review: Methods

Purpose?

• Show that you used scientifically valid methods

• Provide enough info that the reader could repeat the

experiment / study

Turbek et al. (2016) Bull Eco Soc Am

Describe your study organisms & area

Describe your study design

• A reader should be able to replicate it exactly without having

to ask questions!

Describe all of the data analysis you used

• For each statistical result from the Results section, you will have to

describe the analysis method (e.g., ANOVA) in the Methods Section

(we’ll teach you some statistics next week!)

What is a population?

Set of individuals of the same species living in

the same region (a defined area)

Habitat Destruction Urbanization

Agriculture

Deforestation Wildfires & fire suppression

Habitat “Patches”

Image from:

forestfragmentationinethiopia.weebly.com

Areas with a defined configuration that are used

by native species for survival and reproduction

Willamette Valley of Oregon

Fender’s Blue Butterfly

(Icaricia icarioides fenderi)

• Deposit eggs on

leaves in late spring

• Caterpillars eat

leaves over summer

• Caterpillars

overwinter among

leaf litter

• Emerge in March or

April

• Eclose as butterflies

in May

• Adults live ~15 days

Kincaid’s Lupine

(Lupinus sulphureus kincaidii)

• Flowers in May

(when butterflies eclose)

• Seeds “explode” from dried

fruits of this plant

• Only found in prairie habitat

• Threatened by encroaching

invasive plants

Simulation Models • Use field and lab data to create a set of rules that define

outcomes in the system

• Advantage: predict how a system will change if we vary

certain conditions (e.g., temperature, nutrient concentration)

• Disadvantage: often missing knowledge about the system

From Simbio “Patchy Prairies” workbook

Today’s Lab

Learning Goal: use a simulation model

to learn about variables that affect

patch use and population dynamics of

the Fender’s blue butterfly

Experimental Goal: test the efficacy of

different conservation strategies

Exercise 1: Virtual Islands Your TA will lead you through a simulation testing the effect of

patch (“island”) size and distance on butterfly population size.

• Habitat types?

• Why do butterflies head towards an area?

• Why do butterflies move away from an area?

• Turning Probability

• Leave Probability

• What are the 3 conditions for when butterflies

reproduce in the simulation?

• How many offspring does each individual

butterfly have?

• What are the 3 ways that butterflies can die

in the simulation?

Simulation Model’s Parameters

Figure from islandbiogeography.org

Island Biogeography

• Were your predictions based on island biography theory

correct? Why or why not?

• Which habitat configuration had a higher population size?

Exercise 1: Virtual Blues

Large Far Small Near

Edge Effects

Large Far Small Near

Butterflies are more likely to encounter the “edge” of a patch

in Small Near.

Are butterflies more likely to leave the good prairie habitat in

Large Far or Small Near? Why or why not?

• Should patches be large or small?

• Should we have many patches or few patches?

• Should patches be close together or far apart?

What do these results mean for

habit conservation?

Benefit: prevents invasive species

Cost: kills butterflies (larvae in the leaf litter)

Fire

Fires start approximately every 40 (virtual) weeks

Fires occur at same size and rate in both habitat configurations

Fires kill ALL butterflies and lupine plants in the patch

(localized extinction WITHIN a patch)

Exercise 2: Hot & Bothered

• Which habitat configuration had higher population sizes?

• Which habitat configuration had lower extinction rates?

• What aspect of butterfly behavior is making that habitat

configuration better for butterfly survival (i.e., increasing

population stability) during periodic fire disturbance?

Exercise 2: Hot & Bothered

Large Far Small Near

• Should patches be large or small?

• Should we have many patches or few patches?

• Should patches be close together or far apart?

What do these results mean for

habit conservation?

Experimental Design REPLICATE: repeat the same control & experimental

conditions multiple times

Statistics must be calculated from multiple, random samples.

Why?

1. Decrease error

2. Account for variation (individual, population, site, etc.)

3. The more samples that you collect, the closer your

statistical estimate is to the true value of the parameter!

Salamander 1: 27 mm

Salamander 2: 32 mm

Avg: 29.5 mm

True: 28 mm

27 27 28

26 27 29

26 28 31

26 30 27

27 28 25

28 32 28

26 30 26

31

Avg: 27.8 mm

True: 28 mm

CONTROL: a condition used for comparison to determine

the effect of your independent variable on the dependent

variable (a “baseline”)

• No response

• “Normal response”

Experimental Condition:

mango tree given fertilizer

with both nitrogen N & phosphorous P

Controls:

• Mango tree with nothing added

• Mango tree’s normal response to only N

• Mango tree’s normal response to only P

CONTROL:

Experimental Design

Do all ecological studies have controls?

Experimental: Researchers usually manipulate the species or

sites themselves. They must have a control. Some studies

have a “natural” existing control (e.g., non-bulldozed site as a

control for bulldozed site in rainforest).

Observational: observations of one population or

comparisons among groups or populations without a clear /

manipulated control

Example: I can observe and compare the behavior of male and

female Cuban Treefrogs (Osteopilus septentrionalis) and draw

conclusions about differences between sexes; however, no

control exists.

For exercises 1 and 2, were the simulations experimental or

observational?

Habitat Restoration

Corridor

Stepping Stones

Patch

Enlargement

Figure from Science Eaux & Territoires

Research Question 1

Which habitat restoration method improves

the population size and stability of Fender’s

blue butterflies?

Are you conducting an experimental or observational

study?

Is there a control?

Replicates? How many times should our class test each

habitat configuration?

4 groups: current habitat configuration at River Ridges Reserve,

stepping stones, corridor, patch enlargement

Exercise 4: Connections

Current habitat configuration

Stepping Stones

X hectares

Patch Enlargement

X hectares

Corridor

X hectares

1. Select the desired habitat configuration (e.g.,

TZ_Control_1) from My Saved Patches.

2. Open the automator. Set it to 100 weeks and 20 runs.

3. Hide butterflies so that the simulation runs faster.

4. Run the automator and record extinction rate and

population size in the shared spreadsheet.

5. Repeat this process for all of your habitat

configuration versions.

Q1: Data Collection

Research Question 2

Do stepping stones improve population

size and stability during prescribed burns?

Are you conducting an experimental or observational study?

Is there a control?

Replicates? How many times should our class test each

habitat configuration?

2 groups: current habitat configuration at River Ridges Reserve,

stepping stones

1. Select the desired habitat configuration (e.g.,

TZ_Control_1) from My Saved Patches.

2. Select “Periodic Fires” under parameters.

3. Open the automator. Set it to 100 weeks and 20 runs.

4. Hide butterflies so that the simulation runs faster.

5. Run the automator and record extinction rate and

population size in the shared spreadsheet for control

patches.

6. Repeat this process for stepping stones.

Q2: Data Collection

Statistics: R and R Studio

First, download the FREE program R here:

https://cran.cnr.berkeley.edu/

Then, download the FREE version of RStudio here:

https://www.rstudio.com/products/rstudio/download/

Make sure to OPEN Rstudio before the next lab!

The software has to download some items when

you first open it.

If you cannot install Rstudio Desktop, please email

your TA immediately.

Next Class

Assignment 2 (Methods) due on Canvas before

11:59 on Monday September 7

Next lab will be September 15

Quiz 2

• Review today’s lecture & required reading

• Read required reading for “Statistics”

Download R BEFOE next Class