1. Ecological niche and ecological niche modeling
Tereza Jezkova
School of Life Sciences, University of Nevada, Las Vegas
March 2010

2. What drives species distributions?
All species have tolerance limits for environmental factors beyond which individuals cannot survive, grow, or reproduce

3. Tolerance Limits and Optimum Range Tolerance Limits
Environmental Gradient
Tolerance limits exist for all important environmental factors

4. Critical factors and Tolerance Limits

5. Critical factors and Tolerance Limits
For some species, one factor may be most important in regulating a species’ distribution and abundance.
Usually, many factors interact to limit species distribution.

6. ? Critical factors and Tolerance Limits
Organism may have a wide range of tolerance to some factors and a narrow range to other factors

7. Specialist and Generalist species...
Fig. 4-11, Miller & Spoolman 2009

8. FUNDAMENTAL NICHE
Biotic factors
Historical
factors
REALIZED
NICHE
Realized
environment

9. Fundamental versus realized niche
Tolerance Limits and Optimum Range
Fundamental versus realized niche
Fundamental (theoretical) niche
- is the full spectrum of environmental factors that can be potentially utilized by an organism
Realized (actual) niche
- represent a subset of a fundamental niche that the organism can actually utilize restricted by:
- historical factors (dispersal limitations)
- biotic factors (competitors, predators)
- realized environment (existent conditions)

10. Tolerance Limits and Optimum Range Niche shift
Are niches stable?
Realized niche shifts all the time due to
changing biotic interations,
realized environment,
time to disperse
NO!
Time T1
Time T2
Realized
Niche
Shift

11. Fundamental niche shift when tolerance limits change  due to evolutionary adaptation
Time T1
Time T2
Fundamental
Niche
Shift

12. Resource Partitioning
Law of Competitive Exclusion - No two species will occupy the same niche and compete for exactly the same resources
- Extinction of one of them
- Niche Partitioning (spatial, temporal)

13. Niche partitioning and Law of Competitive Exclusion
Chthamalus
Balanus
Chthamalus
Balanus

14. Niche partitioning and Law of Competitive Exclusion

15. Ecological niche modeling
Purpose: ·   
- Approximation of a Species Distribution
    

16. Ecological niche modeling
Purpose: ·  
     - Potential Niche Habitat Modeling
(Invasive species, diseases)
   

17. Ecological niche modeling
Purpose: ·   
    - Site Selection or conservation priority:
Suitability Analysis

18. Ecological niche modeling
Purpose: ·      
- Species Diversity Analysis

19. Ecological niche modeling
Two types:
1. DEDUCTIVE: A priori knowledge about the organism
Example: SWReGAP

20. Ecological niche modeling
Two types:
2. CORRELATIVE: Self-learning algorithms based on known occurrence records and a set of environmental variables

21. Occurrence records:
Own surveys (small scale)
Digital Databases (e.g. museum specimens)
MANIS (mammals)
ORNIS (birds)
HERPNET (reptiles)
HAVE TO BE GEOREFERENCED (must have coordinates)

22. WORLDCLIM http://worldclim.org/
Variables:
Temperature (monthly)
Precipitation (monthly)
19 Bioclimatic variables
Current, Future, Past
Resolution:
ca. 1, 5, 10 km
Coverage
World

23. Southwest Regional Gap Analysis Project http://earth.gis.usu.edu/swgap
Northwest GAP Analysis Project
Variables:
Landcover
Resolution:
ca. 30 m
Coverage
western states

24. Natural Resources Conservation Service (NRCS)
SSURGO Soil Data
Variables:
Soils
Resolution:
ca. 30 m
Coverage
USA but incomplete 

25. Ecological niche modeling
Step 1: occurrence records
Step 2: environmental variables
Step 3: current ecological niche
Step 4: projected ecological niche
How are these predictions made? In general, they use the methodology of ecological niche modeling. This methodology can be described in four steps. First, the researchers assemble occurrence records for the target species – coordinates where the species was documented as “present. Second, they assemble digital maps environmental variables that they suspect could be important in delimiting the species distribution – these environmental variable could be climate (tem and prec), also soils, vegetation, food resources, competition and so on. This simple example would represent temperature along elevation gradient or latitudinal gradient. In the third step, they delimitate suitable habitat based on the information learned from the occurrence records – the green rectangle represent the suitable niche for our hypothetical species. In the fourth step, the same environmental variables are obtained under a scenario of a climate change, in this case global warming, and the range shift is inferred. But this methodology makes one major assumption:

27. Ecological niche modeling – models from Maxent

28. !!! WRONG INTERPRETATIONS !!!
Problems: Models are only as good as the data that goes into it!!!
CALIBRATION MODELS
Insufficient or biased occurrence records
Insufficient or meaningless environmental variables
PROJECTION MODELS
Inaccuracies in climate reconstructions
Dispersal limitations
Non-analogous climates
Niche shift (evolution)
!!! WRONG INTERPRETATIONS !!!

29. sasquatch
blackbear

30. Exercise (work in pairs):
Download museum records for one of nine species
Prepare occurrence data file
Run the program Maxent for current (0K) and last glacial maximum (LGM) climate
Make maps in DivaGIS (or ArcGIS if you have it)
Answer questions on the worksheet
This PowerPoint is on the website, so are the 0K and LGM datasets
Detailed instructions are at the end of this PowerPoint

31. Species:
MAMMALS:
Chisel-toothed kangaroo rat (Dipodomys microps)
Desert kangaroo rat (Dipodomys deserti)
Pygmy rabbit (Brachylagus idahoensis)
Pika (Ochotona princeps)
Mountain beaver (Aplodontia rufa)
REPTILES and AMPHIBIANS:
Desert Horned Lizard (Phrynosoma platyrhinos)
Coastal Tailed Frog (Ascaphus truei)
Long-nosed Leopard Lizard (Gambelia wislizenii)
Gila monster (Heloderma suspectum)

32. Download Occurrence Records
Choose either Manis database (mammals) or Herpnet database (reptiles)
Select “Data portals”
In Manis, click on any of the three providers (e.g. MaNIS Portal at the Museum of Vertebrate Zoology
Click “build query”
Click “Arctos-MVZ catalog” and scroll down
Click on “select a concept” and choose “scientific name”
Click on “select a comparator” and choose “contains (% for wildcard)
Type in the scientific name (e.g. Dipodomys deserti)
Delete number under “Specify record limit”
Click on “submit query”
WAIT !!!
If the server crashes start over again ;)
When the server returns the result of your search, click on “Download tabular results” and save the file into a folder

33. Excel – prepare occurrence records csv. file
Open downloaded occurrence records in Excel (right-click and use the “open with” function
Delete unnecessary rows up front
Sort by “coordinate uncertainty”
Delete all records with no coordinates or those with coordinate uncertainty more than 5000 meters
Delete all columns except the species, latitude and longitude
Make sure the column representing the species has the same value in all cells
Format the columns representing latitude and longitude as numbers with 4 decimal places (Font – Format cells – Number – Number – 4 decimal places)
Save as “ .csv “

34. Maxent
Download the 0K and LGM bioclimatic variables
Unzip each dataset into a separate folder
Open Maxent (*.bat file)
Import your *.csv file of occurrence records
Import the folder with the 0K bioclimatic variables
Check all three fields
Indicate the directory with the LGM
layers
Indicate your output directory
Press “Run”

35. Diva GIS
Import your occurrence records by selecting: Data -> Import points to shapefile -> From text file (.txt)
Add the shapefile representing “states”: Layer –> add layer –> States.shp
(unzip first)
Import your 0K model generated by Maxent (your_species.asc) by selecting: Data -> Import to Gridfile ->Single file. Choose “ESRI ascii” of file and “select integer”
Repeat for your LGM model (your_species_ccsm.asc)
Use the zoom tool to zoom in or out to capture the model well
Unclick the LGM model
Click on “Design” in the bottom right corner and click “OK” in the top left corner
Save as *.bmp file
Click on “data” in the bottom right corner, unclick you OK model and check your LGM model.
Click on Design and repeat your steps as before

36. BIOCLIMATIC VARIABLES
BIO1 = Annual Mean Temperature
BIO2 = Mean Diurnal Range (Mean of monthly (max temp - min temp))
BIO3 = Isothermality (P2/P7) (* 100)
BIO4 = Temperature Seasonality (standard deviation *100)
BIO5 = Max Temperature of Warmest Month
BIO6 = Min Temperature of Coldest Month
BIO7 = Temperature Annual Range (P5-P6)
BIO8 = Mean Temperature of Wettest Quarter
BIO9 = Mean Temperature of Driest Quarter
BIO10 = Mean Temperature of Warmest Quarter
BIO11 = Mean Temperature of Coldest Quarter
BIO12 = Annual Precipitation
BIO13 = Precipitation of Wettest Month
BIO14 = Precipitation of Driest Month
BIO15 = Precipitation Seasonality (Coefficient of Variation)
BIO16 = Precipitation of Wettest Quarter
BIO17 = Precipitation of Driest Quarter
BIO18 = Precipitation of Warmest Quarter
BIO19 = Precipitation of Coldest Quarter