Impact of labyrinth geometric design on emitter clogging using dimensional analysis

Since its widespread adoption in the 1970s, drip irrigation has continued to expand steadily worldwide. There are many reasons explaining this continuous growth, but among them is its theoretical application efficiency, which is around 90%. However, the same cannot be said of its actual performance, which sometimes falls to unacceptable values due to clogging. This study examines the geometry of emitter labyrinths and its influence on clogging with the aim of identifying the optimal combination of hydraulic and geometric variables. To this end, three new lines of research were pursued that involved: (i) defining more precise geometric parameters and new dimensionless numbers based on them; (ii) including hydraulic parameters; and (iii) establishing direct correlations with the most common clogging indices. The test was conducted in an experimental greenhouse using 20 emitters with differing characteristics (hydraulic behavior and insertion). Treated wastewater that had undergone tertiary treatment was used as the irrigation source. In the experiment, three irrigation treatments were established using this reclaimed water: Reclaimed Wastewater (RW), Reclaimed Wastewater + Fertilization with Standard Ionic Solution (RW-FR), and Reclaimed Wastewater + Fertilization with Standard Ionic Solution + Maintenance (RW-FR-M). Through dimensional analysis, correlations were established with four clogging indices: Dra, CU, Us, and DU.The emitters were grouped according to their hydraulic performance (non-compensating and compensating pressure) and insertion type (in-line, inter-line, and on-line). Model fits were checked using the normalized root mean square error (NRMSE), coefficient of determination (R 2 ) and mean absolute error (MAE). For all clogging indices, compensating pressure emitters consistently yielded better values than non- compensating ones. Regarding insertion type, inter-line emitters performed best, followed by in-line and on-line models. The good fits obtained indicate that under the experimental conditions of this study, geometric and hydraulic variables could reasonably predict clogging indicators.