Groundwater level trend analysis using the statistical auto-regressive HARTT method

In this study, the Hydrograph Analysis: Rainfall and Time-Trends (HARTT) model was used to determine the contribution of climatic and non-climatic stresses on groundwater levels in the Lake Haramaya well-field, Ethio‐ pia. Monthly precipitation and monitored water-level data were used as explanatory variables of the method. Variabil‐ ity in rainfall explained 81.3% of groundwater levels using 2-month average time-delay. The coefficient of the impact of rainfall on groundwater level (K1) was found to be 0.00562 ± 0.0007 mm. This K1 value indicates that a 1 mm increase in rainfall from the annual average rainfall raises the groundwater-level by 0.00562 ± 0.0007 mm, while 1 mm decrease in rainfall causes a 0.00562 ± 0.0007 mm drop in groundwater-level in the area. However, the aver‐ age falling trend of the groundwater level (K2) was 1.51 ± 0.133 m/year, even with rainfall causing water-levels to rise between 1.01 to 3.29 m/year. With decreased rainfall, rainfall accounted for about 19.5% of the total-drawdown, while 80.5% was due to cumulative effects of non-climatic variables. This shows that rainfall inputs are negated by cumulative non-climatic stresses leading to the long-term net decline in groundwater level. Projected water-level results show that groundwater levels will be below pump‐ ing positions in <24 years which may have dire conse‐ quences for local landowners.