Get environmental temperatures

From DEBwiki
Jump to: navigation, search
AmP estimation
Concepts

Data and completeness
Parameter estimation
Goodness-of-fit: SMSE / MRE
AmP Literature

Practice - essentials

Starting an estimation for a new species
Setting initial parameter values
Setting weight coefficients
Computing implied properties
Submitting to the collection

Practice - extra modules
Code specification
User-defined files: run, mydata, pars_init, predict
Data: Zero-variate, Univariate, Pseudo-data

Typified models
Estimation options

Background and Motivation

All rates and ages depend on temperature. Temperature is stored as auxilliary data and is usually specified in the mydata file, and used for calculating model predictions.

Correcting for temperature using the Arrhenius temperature correction model amounts to a simple transformation of ages and rates by multiplying or dividing by temperature correction factor (c_T). What is important to keep in mind though is that metabolic rates depend on temperature in a non linear way. If the temperature was fluctuating rather than constant, the mean temperature has a non-straight-forward relationship with the mean c_T.

A number of data sets (field or laboratory) are acquired at non-constant (fluctuating) temperatures. One is tempted to calculate the mean value of temperatures (T_mean) and then correct the metabolic rates with the c_T for that (T_mean) temperature. The problem arises, however, when the experienced (fluctuating) temperatures had a larger effect in low or high part of the range, which is often the case.

The presented procedure shows how to: 1. Obtain environmental temperature values if temperatures are not reported but the geographic location is known; 2. Use the temperature values in combination with the Arrhenius temperature correction model, and 3. Obtain and use c_T as a function of time. For a concrete example including code for R see below.

The final result is either a (single) temperature (for zero-variate data) or a vector of temperatures (for Uni-variate data) that are taking into account species-specific metabolic traits, and are thus more appropriate in the context of energetic modeling.

Method

Get temperature overview.jpg.jpg

Environmental temperature as a function of time

If you do not have data about environmental temperature associated with your dataset, you can use one of these options.

  • Option 2: Use R to query a global microclimate dataset. These estimates are for the middle day of each month, based on long-term average macroclimates.
  • Option 3: If working with a terrestrial ectotherm, you can improve accuracy of your temperature data (obtained via any of the previous steps) by simulating behavior of the species of interest with the NicheMapR microclimate model.

Defining parameters of the Arrhenius temperature correction model

  • Option 1: Use data on (developmental, metabolic, or other) rates at several different temperatures to estimate parameters of the Arrhenius temperature correction model (T_ref, T_A, T_AL, T_AH, T_L, T_H).
  • Option 2: Find the Arrhenius parameters for your species in literature, and input them directly into your script.

c_T as a function of time and using the back-calculated temperature

Using the temperature correction model, calculate the temperature correction (c_T) for each temperature point obtained in Step 1.

For associating a temperature with a zero-variate data (e.g. age at puberty at a given food level) calculate the mean of all c_T values. Then convert the obtained mean(c_T) to the equivalent temperature using again the Arrhenius parameters. The back-calculated temperature is the 'constant temperature equivalent'.

For obtaining temperatures for specific time points in Uni-variate data, you will need to interpolate using the pairs of time-c_T values you have obtained in previous steps. It is possible to save either the c_T values or back calculated temperature values as a .csv file. This can then be called in the mydata file and stored in the auxData structure.

Example

Here you can find an example where we use the sand lizard, Lacerta agilis, which is in the AmP collection and has field growth observations for a number of locations in Russia. The document describes step-by-step the procedure (including NicheMapR)and provides R code you can use and modify.