[R - nlme] non-linear regression Michaelis-Menten model

ss18

New Member
#1
Hello, I am attempting to model Michaelis-Menten uptake kinetics to predict denitrification rates (response) across a gradient of nitrate loading (predictor) in four seasons (June, August, October, and December) using a mixed effects model. The basic models run well in the nls program in R and allow me to predict a Vmax and a K for each season which was a good jumping off point.

However, I need to run a more complex mixed effects model in nlme to account for pseudoreplication within my experiment. Upon doing so, I have run into some error messages that I am having trouble understanding. I am not sure if it is an issue with my actual data configuration or the way I am coding the model. The code and error message are below:

> aug.m1 <- nlme(dnf.formula, fixed = list(NO3trmt~1, Vm~1, k~1), subset = season == "august", random = Vm ~ 1|CORE, start = c(0,31,0.1), data=nlme2)

Error in chol.default((value + t(value))/2) :
the leading minor of order 1 is not positive definite

Thank you for your time and consideration.