Loading a module in a bash script on an HPC cluster

I am submitting the following bash script with qsub on a standard university cluster,

#!/bin/bash
#$ -cwd           # Set the working directory for the job to the current directory
#$ -pe smp 1      # Request 1 core
#$ -l h_rt=1:0:0  # Request 1 hour runtime
#$ -l h_vmem=100M   # Request 1GB RAM
module load xorg-utils/X11R7.7
module load mathematica/13.1.0
math < topaz43 > topaz43master

but the script is returning an error after queuing for a while, with jade4.e374195 reading

/var/opt/sge/node-b00a-001/job_scripts/374195: line 6: module: command not found
/var/opt/sge/node-b00a-001/job_scripts/374195: line 7: module: command not found
/var/opt/sge/node-b00a-001/job_scripts/374195: line 8: math: command not found

The modules usual load properly when I run them on the login nodes, running

module load xorg-utils/X11R7.7
module load mathematica/13.1.0
math

will start Mathematica as usual. What is going on?