-------------------
GENERAL INFORMATION
------------------

Date: 2019-08-29 
Authors: Thomas Lake
	 Ryan Briscoe Runquist (rbriscoe@umn.edu)
	 David Moeller

Contents: Data for Species Distribution Models and Joint Species Distribution Models of Nine Invasive Species in North America

Data: Data Repository for U of M

Date of data collection: 2017-01-01 to 2019-08-01

Funding Sources: Minnesota Invasive Terrestrial Plants and Pests Center (MITPPC)

Recommended citation for the data:
Lake, Thomas, A.; Briscoe Runquist, Ryan, D.; Moeller, David, A.. (2019). Species Distribution Models and Joint Species Distribution Models of Nine Invasive Species in North America. Retrieved from the Data Repository for the University of Minnesota, http://hdl.handle.net/11299/206482.


---------------------
DATA & FILE OVERVIEW
---------------------
see "SDM DRUM Data.xlsx" for a file inventory 


--------------------------
METHODOLOGICAL INFORMATION
--------------------------
#MAXENT Model Data
Name: MaxEnt software for modeling species niches and distributions
Source: https://biodiversityinformatics.amnh.org/open_source/maxent/
Version: 3.4.1

The ENM Eval Model Results.csv file represents the results of each individual model run. These are the results of how changing model parameters influences the output model statistics. The original code from the R ENMEval package (https://github.com/bobmuscarella/ENMeval) has been modified to output additional columns to the autogenerated output file (see the Data-Specific Information section below for more detail)

The species Tif Projection Stack.csv files are essentially subsets of the larger ENMEval Model Results.csv file. From ENM Eval Model Results.csv, top-performing models were selected. The number of TIFs per stack differ as each species had a different number of top-performing models. These top performing model current and future projections were stacked together into MaxEnt Predictions TIF Projection files, and the TIF stacks are documented in the species TIF Projection Stacks.csv files. 


----------------------------
DATA-SPECIFIC INFORMATION  
----------------------------

##### Current and Future Bioclimatic Variables

Dataset: Current Climatic Data
Name: NA_wc2_BioClim19.grd
Number of Files: 2
Source: http://worldclim.org/version2
Resolution: 30s

Dataset: 2050 and 2070 Future Climatic Data
Name: NA_cc45bi50_BioClim19.grd
Number of Files: 20
File Name Explanation: NA (Extent) CC (CCSM4) 45 (RCP) bi (BioClimatic) 50 (2070) BioClim19 (nineteen variables included)
Source: http://worldclim.org/cmip5_30s
GCMs: CCSM4, GFDL-CM3, IPSL-CM5A-LR, MIROC-ESM, MRI-CGCM3
RCP: RCP45
Resolution: 30s

Variables: 
BIO1 = Annual Mean Temperature
BIO2 = Mean Diurnal Range (Mean of monthly (max temp - min temp))
BIO3 = Isothermality (BIO2/BIO7) (* 100)
BIO4 = Temperature Seasonality (standard deviation *100)
BIO5 = Max Temperature of Warmest Month
BIO6 = Min Temperature of Coldest Month
BIO7 = Temperature Annual Range (BIO5-BIO6)
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

For more information about these variables: http://worldclim.org/bioclim


##### Species Coordinate Data

Dataset: Species Longitude, Latitude Coordinate Pairs
Name: xxx_occurrences.csv or [species]_dataset_[date].csv
Number of Files: 9
Sources: 

EDDMapS. 2019. Early Detection & Distribution Mapping System. The University of Georgia - Center for Invasive Species and Ecosystem Health. Available online at http://www.eddmaps.org/; last accessed August 1, 2019.

Global Biodiversity Information Facility (GBIF) Online Database
**The spreadsheets downloaded from GBIF were processed to remove observations with high levels of uncertainty prior to modeling 

Japanese Hops - humulus_japonicus_clean_22June2018.csv
GBIF.org (25 May 2016) GBIF Occurrence Download. https://doi.org/10.15468/dl.wnav0k

Dalmatian Toadflax - dalmatian_toadflax_occurrences.csv
GBIF.org (21 June 2016) GBIF Occurrence Download. https://doi.org/10.15468/dl.aksyj6

Brown Knapweed - brown_knapweed_occurrences.csv
GBIF.org (13 July 2016) GBIF Occurrence Download. https://doi.org/10.15468/dl.mfyuk1

Common Teasel - common_teasel_occurrences.csv
GBIF.org (23 January 2019) GBIF Occurrence Download. https://doi.org/10.15468/dl.cavtwt

Black Swallowwort - black_swallowwort_dataset.csv
GBIF.org (23 January 2019) GBIF Occurrence Download. https://doi.org/10.15468/dl.fu9wnd

Wild Parsnip - wild_parsnip_occurrences.csv
GBIF.org (23 January 2019) GBIF Occurrence Download. https://doi.org/10.15468/dl.cckfvg

Common Tansy - common_tansy_occurrences.csv
GBIF.org (24 January 2019) GBIF Occurrence Download. https://doi.org/10.15468/dl.e5owy4

Narrowleaf Bittercress - narrowleaf_bittercress_dataset_28June2018.csv 
GBIF.org (25 May 2016) GBIF Occurrence Download. https://doi.org/10.15468/dl.nhu5rb

Oriental Bittersweet - oriental_bittersweet_dataset_16May2018.csv
GBIF.org (11 September 2019) GBIF Occurrence Download. https://doi.org/10.15468/dl.po0unh



##### ENM Eval Model Results.csv

Variables:
model number = Index used to enumerate models.
settings = MaxEnt Feature class settings (LQHPT) with Beta Multiplier value (1-4).
features = Maxent Feature class (Linear, Quadratic, Hinge, Product, Threshold).
rm = Regularization (Beta) Multiplier value used to construct models.

train.AUC = Model AUC value generated from training cross-validated dataset. Training dataset is 75% of data. Testing dataset is 25% of data.
avg.test.AUC = Average AUC value of the testing data, calculated from the training data.
var.test.AUC = Variation between four AUC testing datasets.
avg.diff.AUC = Average difference between four AUC testing datasets.
var.diff.AUC = Variation in difference between four AUC testing datasets. 
avg.test.orMTP = Average minimum training presence (MTP) with 0 percent omission.
var.test.orMTP = Variation in minimum training presence (MTP) with 0 percent omission.
avg.test.or10pct = Average minimum training presence (MTP) with 10 percent omission.
var.test.or10pct = Variation in minimum training presence (MTP) with 10 percent omission.

AICc = Akaike information criterion value.
delta.AICc = difference between AICc of a given model and the AICc of the model with the lowest AICc.
w.AIC = Akaike weight divided by the sum of the likelihood values of all models included in a run. Useful for model averageing.

parameters= Number of MaxEnt model parameters used to fit the MaxEnt model.
 
AUC_bin.1 = AUC of testing bin 1, from a 25% training and 75% testing cross-validated dataset. 
AUC_bin.2 = See AUC_bin.1
AUC_bin.3 = See AUC_bin.1
AUC_bin.4 = See AUC_bin.1

diff.AUC_bin.1 = Difference between AUC of testing bin 1 relative to bins 2-4, from a 25% training and 75% testing cross-validated dataset. 
diff.AUC_bin.2 = See diff.AUC_bin.1
diff.AUC_bin.3 = See diff.AUC_bin.1
diff.AUC_bin.4 = See diff.AUC_bin.1

test.or10pct_bin.1 = MTP with 10% omission of testing bin 1, from a 25% training and 75% testing cross-validated dataset. 
test.or10pct_bin.2 = See test.or10pct_bin.1
test.or10pct_bin.3 = See test.or10pct_bin.1
test.or10pct_bin.4 = See test.or10pct_bin.1

test.orMTP_bin.1 = MTP of testing bin 1, from a 25% training and 75% testing cross-validated dataset. 
test.orMTP_bin.2 = See test.orMTP_bin.1
test.orMTP_bin.3 = See test.orMTP_bin.1
test.orMTP_bin.4 = See test.orMTP_bin.1

TSS_bin.1 = TSS of testing bin 1, from a 25% training and 75% testing cross-validated dataset. 
TSS_bin.2 = See TSS_bin.1
TSS_bin.3 = See TSS_bin.1
TSS_bin.4 = See TSS_bin.1

Boyce_bin.1 = Continuous Boyce Index of testing bin 1, from a 25% training and 75% testing cross-validated dataset. 
Boyce_bin.2 = See Boyce_bin.1
Boyce_bin.3 = See Boyce_bin.1
Boyce_bin.4 = See Boyce_bin.1
Average Boyce = Average of four Boyce_bin.[1-4]

tp_bin.1 = True positive rate of testing bin 1, from a 25% training and 75% testing cross-validated dataset. 
tp_bin.2 = See tp_bin.1
tp_bin.3 = See tp_bin.1
tp_bin.4 = See tp_bin.1

fp_bin.1 = False positive rate of testing bin 1, from a 25% training and 75% testing cross-validated dataset. 
fp_bin.2 = See fp_bin.1
fp_bin.3 = See fp_bin.1
fp_bin.4 = See fp_bin.1

fn_bin.1 = False negative rate of testing bin 1, from a 25% training and 75% testing cross-validated dataset. 
fn_bin.2 = See fn_bin.1
fn_bin.3 = See fn_bin.1
fn_bin.4 = See fn_bin.1

tn_bin.1 = True negative rate of testing bin 1, from a 25% training and 75% testing cross-validated dataset. 
tn_bin.2 = See tn_bin.1
tn_bin.3 = See tn_bin.1
tn_bin.4 = See tn_bin.1

Sens Bin 1 = Sensitivity of testing bin 1, from a 25% training and 75% testing cross-validated dataset. 
Sens Bin 2 = See Sens Bin 1
Sens Bin 3 = See Sens Bin 1
Sens Bin 4 = See Sens Bin 1
Average Sens = Average Sensitivity calculated from Sens Bin [1-4]

V57= Model filename [file stem may not be relevant]


#####JSDM Raster Projection Files
See "JSDM file upload metadata.xlsx" in JSDM Raster Projection Files.zip for more information about these files.


##### [species]_[AreaOfAnalysis]_gjamPredictObj.rda
***For more information see: https://cran.r-project.org/web/packages/gjam/gjam.pdf***

Variables:
x = design matrix 

sdList = list of predictive means and standard errors includes yMu, yPe (predictive mean, SE), wMu, wSe = (mean latent states and SEs) 

piList = predictive intervals, only generated if length(y) < 10000, includes yLo, yHi (0.025, 0.975) prediction interval, wLo, wHi (0.025, 0.975) for latent states 

prPresent = n x S matrix of probabilities of presence 

ematrix = effort 

ychains = full prediction chains if FULL = T