Fig. 4: Effect of water quality on filter performance.

a Peak permeance and volumetric capacity normalized by area for general test water (GTW) and challenge test water (CTW). Each data point represents a different filter (one measurement per filter) with 0.375-inch thickness operated under 1-m gravity head. b, c Filter capacity is most susceptible to humic acid, followed by dust and tannic acid (1-cm diameter, 0.375-inch-thick Eastern white pine filters operated under 1-m gravity head; mean ± s.d.; n = 3 different filters) (see “Methods” for experiment details). In b, either humic acid or tannic acid is added to water. In c, water contains 70 mg/L dust or no dust in either 15 mg/L humic acid or tannic acid. d Decrease in filter permeance with CTW (red) and GTW (blue) is well-fitted by the intermediate fouling model (dashed lines). Each graph represents measurements on a single filter under 1-m gravitational head. (i)–(v): Eastern white pine, 1-cm diameter, 0.375-inch thickness. (vi): Ginkgo, 4-cm diameter, 0.375-inch thickness. e Schematic illustrating the deposition of foulant particles (red) on the pit membrane in the intermediate fouling model, which is consistent with foulant deposition observed by SEM in a partially fouled ginkgo filter. Scale bar, 1 μm. f Pre-treatment with cloth and granular activated carbon (GAC) improves the peak flow rates and volumetric capacity of ginkgo filters with GTW and CTW at 1-m gravity head. Inset shows mean ± s.d. of the volumetric capacity; n = 3, 2, and 4 different filters with CTW, CTW + Cloth + GAC, and GTW, respectively.