Using devRate package on T. solanivora (Lepidoptera: Gelechiidae) in Ecuador

T. solanivora dataset

Dataset for T. solanivora was retrieved from Crespo-Perez et al. 2011¹, using Web Plot Digitizer².

rawDevEggs <- matrix(c(10, 0.031, 10, 0.039, 15, 0.047, 15, 0.059, 15.5, 0.066,
  13, 0.072, 16, 0.083, 16, 0.100, 17, 0.100, 20, 0.100, 20, 0.143, 25, 0.171,
  25, 0.200, 30, 0.200, 30, 0.180, 35, 0.001), ncol = 2, byrow = TRUE)

rawDevLarva <- matrix(c(10, 0.010, 10, 0.014, 10, 0.019, 13, 0.034, 15, 0.024,
  15.5, 0.029, 15.5, 0.034, 15.5, 0.039, 17, 0.067, 20, 0.050, 25, 0.076,
  25, 0.056, 30, 0.0003, 35, 0.0002), ncol = 2, byrow = TRUE)

rawDevPupa <- matrix(c(10, 0.001, 10, 0.008, 10, 0.012, 13, 0.044, 15, 0.017,
  15, 0.044, 15.5, 0.039, 16, 0.034, 15.5, 0.037, 16, 0.051, 17, 0.051,
  20, 0.080, 20, 0.092, 25, 0.102, 25, 0.073, 30, 0.005,
  35, 0.0002), ncol = 2, byrow = TRUE)

Finding models in the literature

Before attempting to fit any model to the empirical data, the devRate function “devRateFind” search the database for previous articles fitting which models to the organism, either by Order, Family, or species.

devRateFind(orderSP = "Lepidoptera")
devRateFind(familySP = "Gelechiidae")

Then, the “devRateInfo” function provides additional information on these models and on parameter estimations.

devRateInfo(eq = taylor_81)

Information from the database can be plotted using the “devRatePlotInfo” function.

devRatePlotInfo (eq = taylor_81, sortBy = "ordersp",
  ylim = c(0, 0.20), xlim = c(0, 50))

Fitting models to empirical datasets

The empirical data can be fitted to any model in the database with the “devRateModel” function and empirical data plotted against the model using the “devRatePlot” function.

mEggs <- devRateModel(eq = taylor_81, temp = rawDevEggs[,1], devRate = rawDevEggs[,2],
  startValues = list(Rm = 0.05, Tm = 30, To = 5))
mLarva <- devRateModel(eq = taylor_81, temp = rawDevLarva[,1], devRate = rawDevLarva[,2],
  startValues = list(Rm = 0.05, Tm = 25, To = 5))
mPupa <- devRateModel(eq = taylor_81, temp = rawDevPupa[,1], devRate = rawDevPupa[,2],
  startValues = list(Rm = 0.05, Tm = 30, To = 5))

devRatePlot(eq = taylor_81, nlsDR = mEggs, temp = rawDevEggs[,1], devRate = rawDevEggs[,2],
  pch = 16, ylim = c(0, 0.2))
devRatePlot(eq = taylor_81, nlsDR = mLarva, temp = rawDevLarva[,1], devRate = rawDevLarva[,2],
  pch = 16, ylim = c(0, 0.2))
devRatePlot(eq = taylor_81, nlsDR = mPupa, temp = rawDevPupa[,1], devRate = rawDevPupa[,2],
  pch = 16, ylim = c(0, 0.2))

Stage	model	Rm	Rm_SE	Tm	Tm_SE	To	To_SE
Eggs	Taylor	0.19340	0.017810	25.34180	0.760130	-6.89390	0.768530
Larva	Taylor	0.06838	0.009473	21.40871	0.729058	5.54497	0.865261
Pupa	Taylor	0.09780	0.008700	22.01140	0.455600	-4.75150	0.471500

Forecasting phenologies from empirical temperature

In this example the temperature dataset is built from a Normal distribution of mean 15 and a standard deviation of 1, with a time step of one day. The developmental models used are those previously fitted with the Taylor model for the three stages. We assumed that the average time for female adults to lay eggs was of 1 day. We simulated 500 individuals, with a stochasticity in developmental rate centered on the developmental rate, with a standard deviation of 0.015 (Normal distribution).

forecastTsolanivora <- devRateIBM(
  tempTS = rnorm(n = 100, mean = 15, sd = 1),
  timeStepTS = 1,
  models = list(mEggs, mLarva, mPupa),
  numInd = 500,
  stocha = 0.015,
  timeLayEggs = 1)

devRateIBMPlot(ibm = forecastTsolanivora, typeG = "density")