Citizen Science - Turbidity (Part 3)

In two earlier posts, here and here, the importance of turbidity and how we measure it were discussed. In this post, I will be looking at a novel way to measure turbidity using a Hanna  Checker colorimeter designed to quantify the phosphate content of water.

Photo 1: Hanna Low Range Phosphate Checker and Cuvette

The optical layout of the Hanna Checker is basically the same as a turbidimeter (see here) insomuch as it measures the amount of light transmitted through the sample. It uses a green LED as the light source which is not ideal (white light or near infrared would be better) but you have to work with what you've got.

I did not calibrate the Hanna for turbidity using formazin standards as this is only a preliminary study. The procedure is very simple. Fill one cuvette with either distilled or de-ionised water; this represents 0 NTU. Vigorously shake the sample to be measured and fill the second cuvette (pre-washed with sample). Insert the 0 NTU cuvette and press C1 (to zero the colorimeter). Take the sample cuvette, invert five times to resuspend any sediment, place it in the cell compartment and press C2 briefly. Record the reading on the screen.

A suite of 14 river samples, collected as part of a citizen science project, with turbidities ranging from 7 NTU to 325 NTU were analysed using the Hanna Checker. The Hanna results, along with the measured turbidities by Nephelometry and Secchi Tube, are collated in Table 1.

Table 1: Hanna Phosphate Checker Readings and Measured Turbidities (NTU) using Nephelometry and Secchi Tube 

Hanna Reading	Turbidity (Nephelometer)	Turbidity (Secchi Tube)
0.01	7	<12
0.04	11	<12
0.05	14	15
0.05	19	21
0.11	26	25
0.22	56	50
0.24	41	25
0.40	69	40
0.48	74	40
0.71	134	100
0.76	122	75
1.11	221	190
1.45	228	200
1.82	325	>240

Note 1: Hanna readings are the mean of five measurements

Note 2: In Figure 2, data points with Secchi turbidities <12 NTU  or >240 NTU excluded from the linear regression analysis

Figure 1 is the linear regression plot of Nephelometric Turbidity vs Hanna Reading. The correlation is very good (R² = 0.984) over a wide range of turbidity values (7 NTU to 325 NTU). The practical range is estimated to be 15 - 400 NTU.

Figure 2 is the linear regression plot between Secchi Tube Turbidity (NTU) and Hanna Readings. The correlation is fair (R² = 0.909) over a more limited turbidity range (15 - 200 NTU) due to the lower precision and accuracy of the Secchi Tube method.

The Hanna Checker has a green LED as its light source and is, therefore, expected to show some dependence on the colour of the sample. Early results suggest this may be the case but further work is required. This method shows some promise as a technique for measuring turbidity in natural waters - albeit with a site-specific calibration.