*2.6. Irrigation Tests*

The irrigation tests were carried out in the plots in order to observe the behavior of the water and the volumes of water used in each irrigation event. The process, in general, is as follows: (1) the initial moisture content in the plot was measured; (2) along the length of the border or furrow distances were marked at every 20 or 30 m, which depended on the length of each of the plots; (3) farmers do not use siphons to flood the plots, instead, they open the unlined channel for the water to pass through and lead it to the furrows or borders with the help of a shovel; (4) the time in which the water front reached each of the established marks (advance front) was counted while the water flow was counted at the entrance of the border or furrow using an ultrasonic Doppler effect sensor (FluxSense®) (SEHIDRO Inc., Qro., MEX); and (5) once the water reached the end of the border or furrow, the water inlet was cut off at the beginning and the time in which the recession wave arrived at all the marked points was taken. The irrigation tests were carried out on irrigation lines that varied from 10 to 60 furrows, depending on the expense at the entrance to the canal. The results of the advance and recession times correspond to the average of the furrows evaluated in each irrigation line.

With the data collected from the plots (length, longitudinal slope, apparent density, initial moisture contents and saturation, and flow at the entrance of furrow or border) and the measured times of the advance and recession phase, we proceeded to find the parameters of the Green and Ampt equation (Ks and hf) values that minimized the sum of the squares of the errors in this test. For this stage, we used the nonlinear Levenberg–Marquardt optimization algorithm [21].
