Week 1 Class Notes
Week 1 Class Notes BIOL3863 001
Popular in General Ecology
Popular in Biosystem Engineering
This 4 page Class Notes was uploaded by Ashley on Tuesday September 1, 2015. The Class Notes belongs to BIOL3863 001 at University of Arkansas taught by John Willson, in Summer 2015. Since its upload, it has received 41 views. For similar materials see General Ecology in Biosystem Engineering at University of Arkansas.
Reviews for Week 1 Class Notes
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 09/01/15
WEEK ONE CLASS NOTES GENERAL ECOLOGY o Ecology the scientific study of the distribution and abundance of organisms and the interactions that determine distribution and abundance o How do we DO Ecology Objective is to understand and explain patterns in distribution and abundance 0 Prediction is usually the overall goal 0 Explanations or causes mechanisms can be proximate or ultimate I Proximate explains a trait or process in terms of immediate physiological or environmental factors I Ultimate explains a trait or process in terms of the evolutionary forces acting upon them 0 EX Movement Patterns in Spotted Salamanders I Question Does more rain increase movements of Spotted Salamanders I Hypothesis More rain causes more salamanders to migrate o Observational descriptive studies looking for patterns in unmanipulated systems I Descriptive studies can build correlative evidence but correlation causation 0 Experiments understanding the mechanisms driving the pattern requires manipulation of the system and replication I Experiments are needed to establish causation 0 Models Theoretical often mathematical representations of ecological processes used to generalize results or make predictions 0 Scientific Method In Ecology l quot32353 0 Quei on Observation amp Hypothesis Generation gtl Hypothesis Predictions 14 lEXPerimenta39 de5 9quot Data Collection Experimentation I Data collection and analysis l Comparing predictions and data Unsuccessful Successful predictions predictions StatlStICS Hypothesis Hypothesis rejection confirmation 39Question answered l o Approaches to Ecological Research H39 h lg Descriptive Field Studies Low Natural Experiments not actually experiments Ecological Ability to realism contra amp Field Experiments repllcate Mesocosm Enclosu re Experiments Low Laboratory Experiments High Field Experiment EX 0 LEAP Landuse Effect on Amphibian Populations I X4 wetlands per site I X3 sites South Carolina Maine Missouri Natural Experiments Descriptive studies that take advantage of existing variation to form treatment groups Treatments are not applied randomly Not true experiments cannot establish causality 000 Also not necessarily natural many natural experiments are created by humans just not by the researcher conducting the study Doing Ecology 0 Question Does forest harvesting affect salamander populations 0 Hypothesis Clearcutting forest causes a decline in salamanders I Study 1 Go to the Ozark National Forest and find areas that have been recently clearcut as well as nearby mature forest stands Survey and count salamanders in several 10 x 10 m plots in each forest type I Study 2 Find sites in the National Forest that are planned for clearcutting by the Forest Service Survey and count salamanders in an area before and after it is clearcut I Study 3 Work with the Forest service to find several mature forest plots Randomly assign half of these plots to be clearcut and leave the other half uncut Survey and count salamanders in each area 1 What approach is being used in each of these studies a Study 1 Descriptivenatural expt b Study 2 Descriptivenatural expt beforeafter design c Study 3 Field experiment 2 Which design would I need to use to say that clearcutting m declines in salamander populations a Only study 3 would allow you to conclude that clearcutting causes salamander declines others are consistent with that idea but are correlative not experimental 3 Which design is best a None of these approaches is best experiment has most scientific power all are quite realistic as opposed to a lab study 1 and 2 are much easier and could include larger sample sizes Hierarchical Scales 0 Scale is a critical consideration in any ecological study 0 Use caution when interpreting results beyond the scale of the study Biological Scales Tim Scales Spatial cales Earth Spheres Ecosystems Communities Populations Individuals Simple Experimental Design Treatment a manipulation of a factor of interest away from ambient conditions Control a treatment representative of ambient conditions Experimental Unit entity to which a treatment or treatment combination are applied independently usually randomly Replicate number of experimental units exposed to each treatment or treatment combination EXAMPLE Effects of food supplementation on rattlesnake populations 0 Select 10 forest plots with rattlesnakes o Randomly Select 5 sites to receive supplemental acorns 0 Measure rattlesnake abundance after several years Experimental Unit Forest Plot Treatments 2 No Supplementation Acorn Supplemented Control Treatment No Supplementation Replication 5 per treatment Results Control Acorn Supplemented of snakes 46344 104676 0 Statistics 0 An objective mathematical way of attaching a level of confidence to conclusions that can be drawn 0 Measures of central tendency and variation are combined mathematically to form a statistic that has an associated level of confidence depending on its magnitude 0 Probabilities and Confidence o Null hypothesis usually that the manipulation or factor has no effect on the response variable 0 Probability p value value between 0 and 1 that describes our confidence level in rejecting a null hypotheses when it is actually true Type I error Alpha value predetermined level of confidence you desire usually 005 5 If pvalue is lower than alpha value we reject the null hypothesis and say that a difference is statistically significant 0 Measures of Central Tendency Histogram Symmetric n 5 M a ri h 2 Histogram slrewed to the right in Median Histogram slrewed to the left m MW o Ttests and ANOVA 0 When we are comparing averages mean of two or more treatments we usually display results as bar graphs and use Ttest or ANOVA statistics to evaluate differences a b l l l Mean number of seeds per plant gt Site A Site B Site A Site B
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'