Resistance measurent

I have used this circuit and program:

http://www.circuitbasics.com/arduino-ohm-meter/

With smaller resistors, and the results are acceptable, about 1.5% error.

However I have a project that requires 8 resistance measurements at a time. If they’re above a set value, turn on an led. Nice and simple.
BUT the resistances I’m trying to measure are around the 100M ohm mark. If I use these in the above circuit my results are all over the place. I’m guessing the Arduino doesn’t have enough power? Is there a way I can do this reliably?

Look here.
High impedance resistance requires a buffer.

Measuring eight 100Megohm resistors with an Arduino A/D with buildin muxer is not easy.
Mains hum and crosstalk from the muxer will upset your measurements.
You could try a 10-33Megohm resistor between +5volt and pin, and a 100n ceramic cap from pin to ground.
With the unknown resistance between pin and ground (wiring short and/or shielded).
Don't expect very accurate results. As FathomT said, you might have to use a fet opamp buffer.
Leo..

At 100M the resistance of possibly humide air and surfaces becomes important, as well as EMF and static electricity on wires.

A buffer amplifier is required, and check carefully its input leakage currents. Analog multiplexers can not be used for their leakage currents. Eventually a higher voltage (24-30V) is required for obtaining reasonable values. Instead of analog measurement I'd use a Wheatstone bridge and 2 comparators for each resistor, for a too high and too low indication.

Capacitors also are very nasty or dangerous. Above suggestion of 10M and 100nF results in a RC time constant of 1s and a settling time of about 5s.

DrDiettrich:
Capacitors also are very nasty or dangerous. Above suggestion of 10M and 100nF results in a RC time constant of 1s and a settling time of about 5s.

Aware of that, but OP didn't mention any time requirements.
Leo..

These posts raise the curiosity - why does the OP want to measure such resistors , and why in multiples ?

Measuring moisture of agricultural crops going into store. 8 spears go into the crop, it measures the resistance, they’re retracted and an the crop is stored according to it’s moisture content. It would take about 5s, whereas manually it takes nearer 5 minutes.

So a slow-ish reaction time is not a big problem

You might find some more info if you Google something like "Arduino wood moisture content probe".
Leo..

No, nothing time critcial about it really, doesn really matter if it does all 8 in turn either.