Datalogging Cyclops Rhodamine WT sensor

That looks like a great sensor. Powering the sensor from the Mayfly battery is probably possible, but at 100ma, it’s toward the upper end of the max operating current of the 5-volt boost circuitry, but it looks like you could power it from either the constant 3.3v rail or using the 3.3v switching circuitry. The bigger issue would be if you wanted to run the sensor continuously or if it’s only going to be sampled for short bursts, with longer power-downs between samples. If you power the Mayfly with a big (4000maH or greater) lithium battery or a pack of C-size batteries, you could get several days of continuously-powered operation. If you only use the sensor for a few seconds every few minutes, then you could get by with smaller batteries and a solar panel and unlimited duration. It all depends on the sample time and sleep duration.

As for measuring the output of the sensor, there are a couple different options. One would be to put the sensor output across a resistor divider that drops the 5v to 3.3v (i.e. a 10k and 20k resistor in series across the sensor output, so when the sensor outputs a max of 5v, the voltage between the two resistors is 2/3 of that, which is 3.3v). Or you could add a separate ADC board (like the Adafruit ADS1115 breakout board). Power the breakout board from the 5v switched output of the Mayfly and then it will be able to handle a max input of 5v. But you’d need to connect the breakout to the I2C port of the Mayfly, so you’d need to add some digital level-shifting on those 2 I2C lines to protect the Mayfly from higher voltage of the breakout. I’m sure there are a few other methods, but those are the two quickest and easiest.