*3.4. Groundwater Level and Capillary Rise*

The average depth to groundwater table varied between 0.7 and 2.8 m in the VIS and between 0.6 and 1.3 m in the BIS during 2014–2016 (Figure 8). In the BIS, the soil waterlogging (detected in the BM well) occurred in August and the deepest level (3 m measured in the BD well) was observed in May. The BNF2 well in the VIS recorded the shallowest groundwater level (0.1 m) in August, while the VF1 well measured the deepest groundwater level (3.4 m) in May. The rise in the groundwater table in August most likely resulted from rainfall recharge.

Furthermore, the groundwater level was influenced by nearby streams, reservoirs, and fish ponds. For example, the BU well in the BIS and the VU, BNM and BoN wells in the VIS exhibited stable and relatively shallow groundwater levels due to their proximity to the Bongo reservoir, Vea fish ponds, and streams, even when deep groundwater levels were recorded at other wells (Figure 8). Irrigation events also impacted on the water table. For instance, the groundwater level in the BF1 well in the tomato field increased steadily from the beginning of the dry season and declined from 4 March, 2015, when 2014–2015 dry season irrigation was over. However, the VF1 and BNF2 wells in the VIS in the tomato and leafy vegetable fields, respectively, exhibited rather variable groundwater levels even during the irrigation period and a downward trend after the end of the irrigation period.

The simulated capillary rise into the root-zone of maize was 43–147 mm in 2014, while in the tomato fields it was 18–157 mm in 2014–2015, and 27–263 mm in the 2015–2016 across the irrigation schemes (Figure 9).

**Figure 8.** Elevation of the groundwater table during (**a**) 2014–2015 dry season in Bongo (**b**) 2015–2016 dry season in Bongo (**c**) 2014–2015 dry season in Vea (**d**) 2015–2016 dry season in Vea (**e**) 2015 rainy season in Bongo and (**f**) 2015 rainy season in Vea.

**Figure 9.** Daily groundwater contribution to the root-zone soil moisture of (**a**) maize in 2014; (**b**) tomato in 2014–2015 and (**c**) 2015–2016 computed by AquaCrop.

### *3.5. Traditional Irrigation Scheduling*

The observed GIA for dry season tomato was lower in 2014–2015 than in 2015–2016 in both schemes (Table 4). Across both irrigation schemes and both dry seasons, the overall range of GIA was 21–67 mm per irrigation event and 584–2559 mm per season. The number of irrigation events for tomato ranged between 20 and 29 in both dry seasons. Particularly in the BIS, the irrigation interval in the tomato fields was generally shorter in 2015–2016 than in the previous dry season, owing to the increased availability of water in the Bongo reservoir.

**Table 4.** Observed field-level irrigation practices and water productivity for tomato in the Bongo and Vea irrigation schemes during the dry seasons.


n.d. = not determined.
