Gully formation in the Ethiopian Highlands has been identified as a major source of sediment in water bodies, and results in sever land degradation. Loss of soil from gully erosion reduces agricultural productivity and grazing land availability, and is one of the major causes of reservoir siltation in the Nile Basin. This study was conducted in the 523 ha Debre-Mawi watershed south of Bahir Dar, Ethiopia, where gullies are actively forming in the landscape. Historic gully development in a section of the Debre-Mawi watershed was estimated with semi structured farmer interviews, remotely sensed imagery, and measurements of current gully volumes. Gully formation was assessed by instrumenting the gully and surrounding area to measure water table levels and soil physical properties. Gully formation began in the late 1980's following the removal of indigenous vegetation, leading to an increase in surface and subsurface runoff from the hillsides. A comparison of the gully area, estimated from a 0.58 m resolution QuickBird image, with the current gully area mapped with a GPS, indicated that the total eroded area of the gully increased from 0.65 ha in 2005 to 1.0 ha in 2007 and 1.43 ha in 2008. The gully erosion rate, calculated from cross-sectional transect measurements, between 2007 and 2008 was 530 t ha⁻¹ yr⁻¹ in the 17.4 ha area contributing to the gully, equivalent to over 4 cm soil loss over the contributing area. As a comparison, we also measured rill and interrill erosion rates in a nearby section of the watershed, gully erosion rates were approximately 20 times the measured rill and interrill rates. Depths to the water table measured with piezometers showed that in the actively eroding sections of the gully the water table was above the gully bottom and, in stable gully sections the water table was below the gully bottom during the rainy season. The elevated water table appears to facilitate the slumping of gully walls, which causes the gully to widen and to migrate up the hillside.

More than two billion people and 40% of global agricultural production depend upon unsustainable groundwater extraction. Managed aquifer recharge (MAR), the practice of strategically recharging water to replenish subsurface storage, is an important subbasin scale practice for managing groundwater more sustainably. However, it is not yet reaching its full potential to counterbalance growing global groundwater demand. Agricultural managed aquifer recharge (Ag-MAR) is an emerging method for spreading large volume flows on agricultural lands and has capacity for widespread global implementation. Yet, knowledge gaps, synergies, and tradeoffs in Ag-MAR research still exist. We identify six key system considerations when implementing Ag-MAR: water source, soil and unsaturated zone processes, impact on groundwater, crop system suitability, climate change and impact on greenhouse gas emissions, and social and economic feasibility. We describe the present distribution, need for common terminology, and benefits of Ag-MAR including groundwater storage, increased environmental flows, and domestic wells support. We then outline major gaps, namely, water quality impacts, and crop health and yield. We showcase the multidisciplinary approach needed for communication and coordination of Ag-MAR programs with stakeholders and the public and provide a framework for implementation. Finally, we outline a vision for the path to Ag-MAR implementation. Ag-MAR is an important approach for achieving groundwater sustainability. However, it is one of many necessary solutions and does not offset the need for groundwater conservation. Graphical Abstract

Groundwater aquifers provide an important “insurance” against climate variability. Due to prolonged droughts and/or irrigation demands, groundwater exploitation results in significant groundwater storage depletion. Managed aquifer recharge (MAR) is a promising management practice that intentionally places or retains more water in groundwater aquifers than would otherwise naturally occur. In this study, we examine the possibility of using large irrigated agricultural areas as potential MAR locations (Ag‐MAR). Using the California Central Valley Groundwater‐Surface Water Simulation Model we tested four different agricultural recharge land distributions, two streamflow diversion locations, eight recharge target amounts, and five recharge timings. These scenarios allowed a systematic evaluation of Ag‐MAR on changes in regional, long‐term groundwater storage, streamflow, and groundwater levels. The results show that overall availability of stream water for recharge is critical for Ag‐MAR systems. If stream water availability is limited, longer recharge periods at lower diversion rates allow diverting larger volumes and more efficient recharge compared to shorter diversion periods with higher rates. The recharged stream water increases both groundwater storage and net groundwater contributions to streamflow. During the first decades of Ag‐MAR operation, the diverted water contributed mainly to groundwater storage. After 80 years of Ag‐MAR operation about 34% of the overall diverted water remained in groundwater storage while 66% discharged back to streams, enhancing base flow during months with no recharge diversions. Groundwater level rise is shown to vary with the spatial and temporal distribution of Ag‐MAR. Overall, Ag‐MAR is shown to provide long‐term benefits for water availability, in groundwater and in streams.

Millions of Californians access drinking water via domestic wells, which are vulnerable to drought and unsustainable groundwater management. Groundwater overdraft and the possibility of longer drought duration under climate change threatens domestic well reliability, yet we lack tools to assess the impact of such events. Here, we leverage 943 469 well completion reports and 20 years of groundwater elevation data to develop a spatially-explicit domestic well failure model covering California’s Central Valley. Our model successfully reproduces the spatial distribution of observed domestic well failures during the severe 2012–2016 drought (n = 2027). Next, the impact of longer drought duration (5–8 years) on domestic well failure is evaluated, indicating that if the 2012–2016 drought would have continued into a 6 to 8 year long drought, a total of 4037–5460 to 6538–8056 wells would fail. The same drought duration scenarios with an intervening wet winter in 2017 lead to an average of 498 and 738 fewer well failures. Additionally, we map vulnerable wells at high failure risk and find that they align with clusters of predicted well failures. Lastly, we evaluate how the timing and implementation of different projected groundwater management regimes impact groundwater levels and thus domestic well failure. When historic overdraft persists until 2040, domestic well failures range from 5966 to 10 466 (depending on the historic period considered). When sustainability is achieved progressively between 2020 and 2040, well failures range from 3677 to 6943, and from 1516 to 2513 when groundwater is not allowed to decline after 2020.