fishmove: Prediction of Fish Movement Parameters based on Multiple...
In fishmove: Prediction of Fish Movement Parameters

Description Usage Arguments Details Value Author(s) References See Also Examples

Functions to predict fish movement parameters of leptokurtic fish dispersal, based on multiple regression.

1	fishmove(species=NA,L=NA,AR=NA,SO=6,T=30,interval="confidence",rep=50,seed=NA,...)

`species`	Fish species (scientific name) from `speciesfishmove`, used for prediction of movement parameters. If `species` is set, then values for length and the aspect ratio are taken from `speciesfishmove` and must not be provided by the user. Any additional argument on fish length and/or aspect ratio overwrites these settings. `speciesfishmove` originates in fishbase.org (Froese and Pauly, 2011).
`L`	Fish length (mm), used for prediction of movement parameters. Also a vector of several fish lengths can be provided.
`AR`	Aspect ratio of the caudal fin, used for prediction of movement parameters (Pauly, 1989). A vector of several aspect ratios can be provided.
`SO`	Stream order (Strahler, 1957), used for prediction of movement parameters. The default value for `SO` is 6. A vector of several stream orders can be provided.
`T`	Time step (days), used for prediction of movement parameters. The default value for `new.time` is 30 (days). A vector of several times can be provided.
`interval`	Type of interval calculation. Can be either "confidence" or "prediction". The default for `interval` is "confidence". See `predict.lm`.
`rep`	Number of regression runs to correct for replicates in `datafishmove`. The default value for `rep` is 50.
`seed`	Number (integer) specifying a fixed seed for the random subsampling process (replicates of regression runs). By default this value is set random.
`...`	do not use.

fishmove predicts movement parameters of leptokurtic fish dispersal. The prediction is based on multiple regression on four variables (fish length, aspect ratio of the caudal fin (Pauly, 1989), stream order (Strahler, 1957) and time).

The package 'fishmove' is based on a meta-analyis of heterogeneous fish movement in rivers (see Radinger and Wolter, 2013)

The movement parameters are calculated for a stationary (sigma_stat) and for a mobile component (sigma_mob) of a fish population. These parameters in combination with a value for the share of each component can be used to describe and display leptokurtic dispersal kernels according to:

F(x) = p * (1/(2*pi*sigma_stat^2)^(1/2))*e^(-(x-mu)^2/(2*pi*sigma_stat^2)) + (1-p) * (1/(2*pi*sigma_mob^2)^(1/2))*e^(-(x-mu)^2/(2*pi*sigma_mob^2))

New values are calculated for a fitted mean and its upper and lower bound (confidence or prediction interval) based on a set of new input variables. In addition, common fish length and aspect ratio, are already provided for appr. 40 different species from fishbase.org (Froese and Pauly, 2011).

`coef.fishmove`	Array of regression parameters of the original multiple regression
`pred.fishmove`	predicted value, upper and lower bound for sigma_stat (movement parameter of stationary component) and sigma_mob (movement parameter of mobile component)

Johannes Radinger

Froese, R. and Pauly D. (2011) FishBase. World Wide Web electronic publication. www.fishbase.org, version (12/2011).

Pauly, D. (1989) A simple index of metabolic level in fishes. Fishbyte, Newsletter of the Network of Tropical Fisheries Scientists 7, 22.

Radinger, J. and Wolter C. (2014) Patterns and predictors of fish dispersal in rivers. Fish and Fisheries. 15:456-473. DOI: http://dx.doi.org/10.1111/faf.12028.

Strahler, A. N. (1957) Quantitative analysis of watershed geomorphology. Transactions of the American Geophysical Union 8, 913-920.

pdk, lm, predict.lm

	# Prediction for a selected fish species with time=365 days
	fishmove(species="Salmo trutta fario",T=365)

	# Prediction with fish length=350 mm, aspect ratio=2, stream order=4 and prediction interval
	# fishmove(L=350,AR=2,SO=4,T=365,interval="prediction")

Loading required package: ggplot2
Loading required package: plyr
Loading required package: MASS
Loading required package: boot
Predicted movement for selected parameters:
, , L=200, AR=1.25, SO=6, T=365

    sigma_stat sigma_mob
fit  106.49849  2590.035
lwr   59.79296  1658.111
upr  189.69572  4046.052