Introduction
Maize (Zea mays L.) is the most extensively cultivated staple crop, playing a key role in ensuring food security across sub-Saharan African nations (Mutuku et al., Reference Mutuku, Roobroeck, Vanlauwe, Boeckx and Cornelis2020). Ethiopia has the largest diversity of maize cultivars, accounting for 43% of the total, and has achieved the second-highest average maize yield in sub-Saharan Africa, averaging about 3.4 Mg ha–1 (Abate et al., Reference Abate, Fisher, Abdoulaye, Kassie, Lunduka, Marenya and Asnake2017). Nationally, maize cultivation covers approximately 2 million hectares, making it the second-largest production area after teff. Approximately 9 million smallholder farmers contribute to 95% of the country’s maize production (CSA, 2021). An estimated 88% of the maize produced is consumed as food, both in its green and dry grain forms (Abate et al., Reference Abate, Fisher, Abdoulaye, Kassie, Lunduka, Marenya and Asnake2017).
Despite its significant role in ensuring food security, the average national maize productivity in Ethiopia falls well below the global average of 5.75 Mg ha–1 (FAOSTAT, 2020). Factors such as water scarcity and declining soil fertility could be contributing to this discrepancy (Baumhardt et al., Reference Baumhardt, Schwartz, Howell, Evett and Colaizzi2013; Yaseen et al., Reference Yaseen, Shafi, Ahmad, Rana, Salim and Qaisrani2014). Water scarcity poses a major challenge to crop production during the long dry season (Teshome et al., Reference Teshome, Bayabil, Schaffer, Ampatzidis, Hoogenboom and Singh2023; Attia et al., Reference Attia, El-Hendawy, Al-Suhaibani, Alotaibi, Tahir and Kamal2021; Greaves and Wang, Reference Greaves and Wang2017). Enhancing irrigation agriculture is vital to bolstering food security, achieving self-sufficiency, increasing food production, and boosting export earnings (Comas et al., Reference Comas, Trout, DeJonge, Zhang and Gleason2019; Baumhardt et al., Reference Baumhardt, Schwartz, Howell, Evett and Colaizzi2013; Gheysari et al., Reference Gheysari, Sadeghi, Loescher, Amiri, Zareian, Majidi, Asgarinia and Payero2017). However, inadequate water supply during the prolonged dry season and escalating competition among water users are significant hurdles in irrigation schemes (Yuana et al., Reference Yuana, Fenga, Huo and Ji2019; Yufenga et al., Reference Yufenga, Saddique, Ajazd, Jiatuna, Khan, Mua, Azmat, Cai and Siddique2021; Zhang et al., Reference Zhang, Ma, Douglas-Mankin, Han and Trout2021). Moreover, the limited water supply is often inefficiently utilized for crop production (Zhang et al., Reference Zhang, Ma, Douglas-Mankin, Han and Trout2021; Jia et al., Reference Jia, Sun, Ali, Liu, Zhang, Ren, Zhang and Jia2016; Teshome et al., Reference Teshome, Bayabil, Schaffer, Ampatzidis, Hoogenboom and Singh2023). In Ethiopia, farmers commonly practice small-scale furrow irrigation, but they frequently apply excessive water to crops, potentially leaching nutrients from the root zone (Asmamaw et al., Reference Asmamaw, Janssens, Dessie, Tilahun, Adgo, Nyssen, Walraevens and Cornelis2021b) and exposing crops to diseases associated with poor aeration, resulting in lower grain quality (Asmamaw et al., Reference Asmamaw, Janssens, Dessie, Tilahun, Adgo, Nyssen, Walraevens, Assaye, Nigate and Cornelis2023). Conversely, farmers may also subject crops to water stress relative to their water requirements, leading to significant yield reductions (Asmamaw et al., Reference Asmamaw, Janssens, Dessie, Tilahun, Adgo, Nyssen, Walraevens and Cornelis2021b). Asmamaw et al. (Reference Asmamaw, Janssens, Dessie, Tilahun, Adgo, Nyssen, Walraevens, Assaye, Nigate and Cornelis2023) corroborated these findings through deficit irrigation field experiments on wheat in Ethiopia. This underscores the imperative of optimizing available water resources by carefully considering the quantity and frequency of irrigation applied within specific crop growing seasons.
Simultaneously, addressing the increasing demand for food entails boosting crop production while minimizing water use (Igbadun et al., Reference Igbadun, Salim, Tarimo and Mahoo2008; Payero et al., Reference Payero, Tarkalson, Irmak, Davison and Petersen2009). Implementing advanced irrigation techniques that deliver precise amounts of water to crops will be pivotal in enhancing water productivity (WP) in future agricultural endeavors (Teshome et al., Reference Teshome, Bayabil, Schaffer, Ampatzidis, Hoogenboom and Singh2023). Consequently, augmenting WP appears as a crucial strategy for ensuring food security (Salemi et al., Reference Salemi, Soom, Lee, Yusoff and Ahmad2011), enabling greater food production to sustain the expanding global population (Yufenga et al., Reference Yufenga, Saddique, Ajazd, Jiatuna, Khan, Mua, Azmat, Cai and Siddique2021). Effective irrigation scheduling, such as deficit irrigation (DI), stands out as a water-saving approach that enhances WP by providing irrigation water below the total crop water requirement (Bayabil et al., Reference Bayabil, Teshome, Guzman and Schaffer2023), potentially yielding higher economic returns by optimizing water use per unit of yield (Asmamaw et al., Reference Asmamaw, Janssens, Dessie, Tilahun, Adgo, Nyssen, Walraevens, Assaye, Nigate and Cornelis2023). Maintaining a water supply level between 60% and 100% of full crop evapotranspiration can improve WP while minimizing yield losses (Fereres and Soriano, Reference Fereres and Soriano2006). However, successful execution requires detailed knowledge of crop water requirements and yield responses, given the variability in crop responses to soil moisture (Comas et al., Reference Comas, Trout, DeJonge, Zhang and Gleason2019; Zhang et al., Reference Zhang, Han, Comas, DeJonge, Gleason, Trout and Ma2019) and soil nutrients (Kibet et al., Reference Kibet, Mugwe, Korir, Mucheru-Muna, Ngetich and Mugendi2023). Therefore, understanding the effect of DI on maize yield and WP is vital for making informed decisions regarding irrigation water management strategies. Previous research has demonstrated that DI can effectively enhance WP across various crops without significantly compromising yield (Teshome et al., Reference Teshome, Bayabil, Schaffer, Ampatzidis, Hoogenboom and Singh2023).
Alongside water scarcity, soil fertility degradation caused by acidity and nutrient depletion stands out as a primary constraint on crop productivity (Wakwoya et al., Reference Wakwoya, Woldeyohannis and Yimamu2022; Agegnehu et al., Reference Agegnehu, Amede, Desta, Erkossa, Legesse, Gashaw, Rooyen, Harawa, Degefu, Mekonnen and Schulz2023). While the severity of this problem varies, soil acidity is a widespread concern across all continents. Approximately 22% of Africa’s tropical and subtropical arable land is affected by soil acidity (Kibet et al., Reference Kibet, Mugwe, Korir, Mucheru-Muna, Ngetich and Mugendi2023). Similarly, in Ethiopia, around 43% of arable land is affected by soil acidity (Asmamaw et al., Reference Asmamaw, Janssens, Dessie, Tilahun, Adgo, Nyssen, Walraevens, Pue De, Yenehun, Nigate, Sewale and Cornelis2022). Soil acidity reduces nutrient availability, leading to aluminum and manganese toxicity (Mosharrof et al., Reference Mosharrof, Uddin, Sulaiman, Mia, Shamsuzzaman and Haque2021). Therefore, adopting integrated management approaches for acidic soils is imperative to tackle existing challenges in crop production. Soil acidity primarily constrains crop production by inhibiting root growth and diminishing the uptake of nutrients and water (Agegnehu et al., Reference Agegnehu, Amede, Desta, Erkossa, Legesse, Gashaw, Rooyen, Harawa, Degefu, Mekonnen and Schulz2023). Acidic soils generate toxicity within the soil solution, impeding plant growth and resulting in reduced crop yields (Kibet et al., Reference Kibet, Mugwe, Korir, Mucheru-Muna, Ngetich and Mugendi2023; Kimiti et al., Reference Kimiti, Mucheru-Muna, Mugwe, Ngetich, Kiboi and Mugendi2021). This toxicity also inhibits microbial activity (Asmamaw et al., Reference Asmamaw, Janssens, Dessie, Tilahun, Adgo, Nyssen, Walraevens, Pue De, Yenehun, Nigate, Sewale and Cornelis2022; Dominguez-Hernandez et al., Reference Dominguez-Hernandez, Zepeda-Bautista, Dominguez-Hernandez, Valderrama-Bravo and Hernández-Simón2019). Crops grown in acidic soils tend to exhibit stunted growth, increased susceptibility to drought, and reduced responsiveness to fertilizer applications, ultimately leading to reduced productivity (Agegnehu et al., Reference Agegnehu, Amede, Desta, Erkossa, Legesse, Gashaw, Rooyen, Harawa, Degefu, Mekonnen and Schulz2023). In Ethiopia, farmers often apply excessive amounts of fertilizer due to the low soil response to crops (Asmamaw et al., Reference Asmamaw, Janssens, Dessie, Tilahun, Adgo, Nyssen, Walraevens, Fentie and Cornelis2021a).
Previous research has demonstrated that both inorganic and organic amendments, including lime, compost, manure, biochar, and inorganic fertilizers, can mitigate the detrimental impacts of soil acidity (Agegnehu et al., Reference Agegnehu, Amede, Desta, Erkossa, Legesse, Gashaw, Rooyen, Harawa, Degefu, Mekonnen and Schulz2023; Kibet et al., Reference Kibet, Mugwe, Korir, Mucheru-Muna, Ngetich and Mugendi2023). In addressing the soil acidity problem, lime is renowned for its significant corrective effects on acidic soils, and is often termed the ‘workhorse’ (Agegnehu et al., Reference Agegnehu, Amede, Desta, Erkossa, Legesse, Gashaw, Rooyen, Harawa, Degefu, Mekonnen and Schulz2023). Liming elevates soil pH levels, enhances the availability of phosphorus and nitrogen nutrients, and reduces exchangeable acidity (Kimiti et al., Reference Kimiti, Mucheru-Muna, Mugwe, Ngetich, Kiboi and Mugendi2021). Additionally, lime directly enhances soil aggregation and structure through the flocculation and cementing actions of calcium carbonate (Dominguez-Hernandez et al., Reference Dominguez-Hernandez, Zepeda-Bautista, Dominguez-Hernandez, Valderrama-Bravo and Hernández-Simón2019). Elevated ionic strength and calcium carbonate concentrations can compress the double layer, thereby reducing repulsion between soil particles. Moreover, liming acts as an amorphous cementing agent, physically binding adjacent particles together to form aggregates (Kibet et al., Reference Kibet, Mugwe, Korir, Mucheru-Muna, Ngetich and Mugendi2023). These flocculating and coagulating effects of liming ultimately lead to improvements in soil physical conditions (Asmamaw et al., Reference Asmamaw, Janssens, Dessie, Tilahun, Adgo, Nyssen, Walraevens, Pue De, Yenehun, Nigate, Sewale and Cornelis2022). However, the cost and transportation of bulk liming materials pose significant barriers for many smallholder farmers in Ethiopia (Agegnehu et al., Reference Agegnehu, Amede, Desta, Erkossa, Legesse, Gashaw, Rooyen, Harawa, Degefu, Mekonnen and Schulz2023; Kisinyo et al., Reference Kisinyo, Opala, Palapala, Gudu, Othieno and Ouma2015). Consequently, there is a need for research into integrated strategies for ameliorating acidic soils to enhance soil quality and increase yields.
The utilization of organic fertilizers enhances both the quantity and quality of soil organic matter, thereby contributing to improved soil quality and providing essential nutrients for subsequent harvests (Thakur et al., Reference Thakur, Sharma, Sankhyan and Kumar2020). Recent research has shown that the residual effects of manure and compost applications significantly elevated electrical conductivity, pH levels, and the availability of phosphorus (Amede et al., Reference Amede, Legesse, Agegnehu, Gashaw, Degefu, Desta, Mekonnen, Schulz and Thorne2021; Kibet et al., Reference Kibet, Mugwe, Korir, Mucheru-Muna, Ngetich and Mugendi2023). Soil microorganisms produce various organic compounds during the decomposition of organic materials (Opala et al., Reference Opala, Odhiambo and Anzetse2023). Soluble humic molecules and low molecular weight aliphatic acids are particularly effective in detoxifying aluminum ions by forming diverse compounds (Haynes and Mokolobate, Reference Haynes and Mokolobate2001). It is crucial to explore alternative strategies for managing acidic soils that take into account the purchasing power of smallholder farmers (Agegnehu et al., Reference Agegnehu, Amede, Desta, Erkossa, Legesse, Gashaw, Rooyen, Harawa, Degefu, Mekonnen and Schulz2023; Asmamaw et al., Reference Asmamaw, Janssens, Dessie, Tilahun, Adgo, Nyssen, Walraevens, Assaye, Nigate and Cornelis2023; Kisinyo et al., Reference Kisinyo, Opala, Palapala, Gudu, Othieno and Ouma2015). Integrated soil fertility management (ISFM) and improved irrigation techniques have the potential to enhance maize productivity and potentially alleviate water scarcity issues. The introduction of DI could be considered as it may alleviate pressure on water resources and expand the areas suitable for irrigation. Therefore, it is crucial to understand the potential impact of combining DI and ISFM on maize production and WP.
Prior studies have primarily focused on investigating how lime, manure, and inorganic fertilizers affect soil characteristics and crop yields (Wakwoya et al., Reference Wakwoya, Woldeyohannis and Yimamu2022). However, there has been no investigation into the combined effects of ISFM and DI on maize yield and WP in acidic Nitisols. Enhancing soil fertility has the potential to increase the WP of crops and optimize the use of available water. Similarly, the availability of nutrients and their absorption by plants greatly depend on adequate soil moisture and its accessibility under low tension. Irrigation and fertilization exhibit synergistic effects on crop yield, emphasizing the importance of their combined application in crop production. Therefore, for irrigation to be effective and advantageous in crop cultivation, the optimal use of nutrients by plants is essential. To address these knowledge gaps, field experiments were conducted over two consecutive irrigated seasons, maintaining normal crop rotation with wheat, to investigate the combined effects of varying rates of lime application, a fixed rate of manure, and inorganic fertilizer in conjunction with DI on maize yield and WP. The specific objectives of this study were to: (i) evaluate the impact of DI and ISFM on maize grain yield and biomass production; (ii) assess the synergistic effects of ISFM and DI on maize grain yield and WP compared to the application of inorganic fertilizer and full irrigation; (iii) evaluate the impact of DI and ISFM on soil water content along the crop growth stages. We hypothesized that liming and manuring would improve soil water retention, leading to an increase in maize grain yield, biomass production, and WP. Additionally, DI when combined with ISFM will result in higher WP compared to full irrigation without ISFM strategies.
Materials and methods
Description of the study site
The field experiment took place within the Koga Irrigation Scheme (KIS) at Ambomesk irrigation block, situated south of Lake Tana in the Upper Blue Nile Basin. The location spans from 12°20´ to 12°31´N and 37°02´ to 37°08´E, at an elevation of 1960 m a. s. l. The Koga Dam has a design capacity of 83 million cubic meters. Water is harvested during the rainy season (late June to the end of August) by diverting the Koga River into the dam. However, from September to the end of May, the river flows naturally, so there is no incoming water to the dam. The KIS is one of the most recent large-scale irrigation projects for farmers. It uses its 1,750 ha reservoir to provide irrigation for nearly 5,828 ha out of a planned 7,004 ha during the dry season, benefiting about 10 356 people (Asmamaw et al., Reference Asmamaw, Janssens, Dessie, Tilahun, Adgo, Nyssen, Walraevens, Fentie and Cornelis2021). The reservoir feeds 12 irrigation blocks and 22 night storage reservoirs. Each block has a secondary lined canal with a total length of 42 km. These secondary canals, fed by the main canal and night storage reservoirs, deliver water to the individual command areas through tertiary and quaternary canals. Surface drainage canals, constructed along one side (either the right or left) of the quaternary canals, are 2 m wide and 0.8 m deep. Additionally, farmers have constructed canals around the edges of their land. These canals serve two purposes: to prevent salinity during the irrigation season and to safely drain excess water to natural waterways during the heavy rain season. In the future, more drainage canals may be necessary to protect against salinity as groundwater levels rise. The KIS is characterized by a humid to sub-humid climate and experiences an average air temperature of 24°C, and an average annual rainfall of 1,528 mm (Asmamaw et al., Reference Asmamaw, Janssens, Dessie, Tilahun, Adgo, Nyssen, Walraevens and Cornelis2021b). Rainfall distribution is unimodal. The rainy season generally begins in June, occasionally at the end of May, and typically ends by mid-September. During the maize-growing season, weather data, including daily rainfall, maximum and minimum temperatures, relative humidity, sunshine hours, and wind speed (measured at a height of 2 m), were collected from the Bahir Dar meteorological station, located 35 km north of the experimental field. During the irrigated maize-growing seasons, no rainfall occurred, except for a single event at the end of May in both seasons (Table 1).
Average weather conditions during irrigated maize growing seasons (2019 and 2020)
Where, Tmax is maximum temperature, Tmin is minimum temperature, RH is mean relative humidity, ETo is reference evapotranspiration.
Soil analyses following four irrigated and two rain-fed cropping seasons treatments are presented in Table 2. The soil in KIS exhibits low sand content and high clay content, typical of clayey Nitisols. The soil pH under the reference treatment (C) indicates strong acidity, possibly due to the leaching of water-soluble exchangeable cations because of over-irrigation and heavy rainfall during the rainy season (Asmamaw et al., Reference Asmamaw, Janssens, Dessie, Tilahun, Adgo, Nyssen, Walraevens, Pue De, Yenehun, Nigate, Sewale and Cornelis2022). However, the application of lime at varying doses (L1, L2, L3) and manure (M) resulted in increased soil pH and reduced bulk density compared to C. ISFM significantly influenced water retention at field capacity and saturated hydraulic conductivity.
Basic hydro-physical soil properties under integrated soil fertility management strategies after four irrigated and two rain-fed cropping seasons treatment applications. Standard deviations are presented in parenthesis
Where, FC and PWP are volumetric water content at field capacity (–33 kPa) and permanent wilting point (–1500 kPa), respectively, BD is bulk density. L1 is a combination of 1.43 t ha–1 of lime with 3 Mg ha–1 of manure and full doses of urea (200 kg ha–1) plus 200 kg ha–1 of NPSB (Nitrogen, Phosphorus, Sulphur and Boron), hereafter referred to as inorganic fertilizer; L2 is a combination of 1.15 Mg ha–1 of lime with 3 Mg ha–1 of manure and full doses of inorganic fertilizer; L3 is a combination of 0.86 Mg ha–1 of lime with 3 Mg ha–1 of manure and full doses of inorganic fertilizer; M is a combination of 3 Mg ha–1 of manure along with full doses of inorganic fertilizer; and C is using only full doses of inorganic fertilizer.
Treatments and experimental design
DI and ISFM experiments were investigated over four consecutive seasons (2018 to 2020). Following the common practice of local farmers in the study area, a cropping sequence of wheat (Triticum aestivum L.) followed by maize (Zea mays L.) was done, with wheat cultivated from October to January and maize from January to May. But here, only the maize results are presented.
The effects of DI and ISFM on maize production and WP were assessed through a full factorial experiment comprising five ISFM treatments (Table 3) and four irrigation water treatments, each replicated three times. The DI and ISFM treatments were randomly distributed across three blocks (Supplementary Material Fig. S1), with each block containing a complete replication of the treatment arrangement. Uniform conditions were maintained across all plots, including soil conditions, field management practices (such as plowing, weed, and pest control), land slope (leveling), plot sizes and shapes, and furrow lengths, except for variations in DI and ISFM treatments. Furthermore, field management activities were carried out simultaneously across all plots.
Integrated soil fertility management treatments investigated in Koga irrigation scheme in 2019 and 2020
Where, NPSB is nitrogen, phosphorus, sulphur, and boron, 0 indicates no lime or organic input. The ISFM treatments comprised the following: (i) incorporation of 1.43 Mg ha–1 of lime (representing 100% of lime requirement of the soil (LR)) with 3 Mg ha–1 of manure and full-dose inorganic fertilizer (L1); (ii) integration of 1.15 Mg ha–1 of lime (representing 80% of LR) with 3 Mg ha–1 of manure and full-dose inorganic fertilizer (L2); (iii) incorporation of 0.86 Mg ha–1 of lime (representing 60% of the lime requirement, LR) with 3 Mg ha–1 of manure and full-dose inorganic fertilizer (L3); (iv) integration of 3 Mg ha–1 of manure with full-dose inorganic fertilizer (M); (v) a control treatment using full-dose urea and NPSB (nitrogen, phosphorus, sulfur, and boron), referred to as inorganic fertilizer (C). Each experimental plot measured 8 m in width and 30 m in length, with a 2 m untilled buffer separating each plot and block from one another on all sides.
Three deficit irrigation scenarios, representing 80%, 60%, and 50% ETc, along with full irrigation (100% ETc), were evaluated. Full irrigation served as the control for the DI scenarios. ETc represents crop evapotranspiration, indicating the crop’s water requirement (i.e., the product of reference evapotranspiration and crop coefficient), as described in section 2.4.
Soil fertility and crop management
A well-decomposed organic manure was applied at a rate of 3 Mg ha–1 in the first round of irrigation season experiments in 2018. Farmers manually spread and incorporated the manure into the soil using an ox-driven Maresha plow, reaching a depth of 10 to 15 cm, on the same day for all treatments except the control, which did not receive any manure. Similar to the manure application, lime was incorporated into the soil to a depth of 10 to 15 cm approximately one month prior to planting, under dry soil conditions. Inorganic fertilizers containing urea (46% N, 0% P, and 0% K) and NPSB were applied. A full dose of urea (200 kg ha–1) was uniformly applied for all treatments. The urea was split into two applications: 50% was applied 35 days after sowing, with the remaining 50% applied at the tasselling stage. Additionally, NPSB, containing 18.9% N, 37.7% P, 6.95% S, and 0.1% B, was applied at a full dose (200 kg ha–1) at the time of sowing.
The hybrid maize variety P3812W, sourced from Pioneer Seed Company, and known for its high yield, early maturation, and moderate drought tolerance, was used for planting. Sowing was done manually by drilling in rows, following thorough land preparation, with a seeding rate of 25 kg ha–1 on February 1, 2019, and February 7, 2020. Seeds were planted at a spacing of 75 cm between rows and 25 cm between plants within rows. Weed control was managed manually, and the crop was regularly monitored for pests and diseases. The identification of maize phenological stages was achieved by observing the plants’ growth and development and comparing these observations to established growth stage classifications, such as the BBCH scale (Biologische Bundesanstalt, Bundessortenamt, and CHemical industry scale). This process involved regular field observation practices, including weekly inspections of maize plants to identify transitions between stages. Additionally, we recorded the dates of key events, such as emergence, leaf development, tasseling, silking, and maturity, to effectively track crop progress. Harvesting took place on May 24, 2019, and May 29, 2020, respectively.
Crop water requirement and irrigation water management
The seasonal crop water requirements and irrigation scheduling for maize during the experimental seasons are presented in Table 4. Daily reference evapotranspiration was computed using CropWat program, employing the FAO Penman-Monteith equation as described by Allen et al. (Reference Allen, Pereira, Raes and Smith1998). Weather data were gathered from the National Meteorological Service Agency Bahir Dar branch. Root depth, crop coefficient (Kc), and optimal soil moisture depletion level were derived from FAO Irrigation and Drainage Paper 56 (Allen et al., Reference Allen, Pruitt, Wright, Howell, Ventura, Snyder, Itenfisu, Steduto, Berengena, Yrisarry, Smith, Pereira, Raes, Perrier, Alves, Walter and Elliott2006). The Kc values were used to calculate the crop water requirements and then to design the irrigation schedule. The water stress threshold (maximum allowable depletion, MAD) for maize was used to assess the water stress levels at various phenological stages.
Seasonal crop water requirement and irrigation scheduling for 2019 to 2020 generated from CropWat program version 8.0
Where NIW = net irrigation water amount, AIW = applied irrigation water amount, Kc = crop coefficient, ETc = seasonal crop water requirement, Init, Dev, Mid and Late = initial, development, mid and late wheat-growing stages, respectively.
We investigated the effects of reducing the net irrigation requirement (100% ETc) by 50%, 40%, 20%, and 0% of ETc, throughout the entire maize growing season. A rain-fed condition (100% reduction) was not considered due to the likelihood of complete crop failure. Furrow irrigation was employed, with irrigation sessions occurring every 10 days. Discharge measurement was facilitated using a calibrated standard Parshall flume with a 15-cm throat width installed at the inlet of the main plot (Supplementary Material Fig. S2). Water was channeled from a quaternary canal (unlined) to the field plot at a constant discharge rate, with the discharge then directed into each plot and furrow at predetermined times. The time needed to achieve the desired water depth was calculated by multiplying the depth of water needed for each irrigation session (m) by the area of the plot to be irrigated (m2), then dividing by the discharge (m3 s−1). The discharge was immediately stopped by shutting the channel banks to prevent water from entering the plots after the necessary depth had been applied to a particular plot. The discharge was determined by multiplying the flume discharge constant (coefficient) with the depth measured at the Parshall flume (in m) and the discharge exponent.
Soil moisture data collection
Gravimetric soil water content (SWC) was assessed one day prior to and two days subsequent to each irrigation session at depths ranging from 0–20 cm, 20–40 cm, 40–60 cm, and 60–100 cm, with three replicates for each treatment, considering the maize root zone. Samples were collected from multiple locations within the ridge of each plot during every irrigation event over the study periods. The average SWC measured across the 0–100 cm soil layer were employed to determine the mean SWC within the soil profile. Soil samples were obtained using 10 cm height core samplers, and subsequently oven-dried at 105°C for a duration of 48 hours.
Agronomic data collection, yield, and biomass quantification
Grain yield and aboveground biomass data were gathered during harvest from a sampled area of 9 m2 within each plot, with three replicates taken into consideration for statistical analysis across all treatments. Following detachment of maize cobs and grains from the stubble, and sun-drying, grain yield weight was measured. Grain yield was calculated as the weight of the harvested grain and adjusted to the standard moisture content of 12.5% for maize, and subsequently converted to Mg ha–1. To address border effects, the outermost rows were excluded from sampling. For biomass estimation, stubbles were carefully cut approximately 2 cm above the surface. The total fresh weight of maize cobs and biomass was measured in the field. Twelve stubbles of varying sizes were randomly selected from the measured biomass, cut into smaller pieces, well-mixed, and subsampled. The fresh weight of the subsampled biomass and twelve maize cobs of different sizes were recorded. Subsequently, the subsamples were dried in an oven at 65°C for 48 hours, and the dry weight was recorded.
Grain yield was also expressed per unit of net irrigation water requirement (NIW) for evaluating the potential yield under DI when using the same amount of water as with full irrigation but on land currently not irrigated. Yield per total NIW was estimated by multiplying the yield per hectare for a specific ISFM treatment by a factor of 1.25, 1.67, and 1.99 for 80%, 60%, and 50% of crop evapotranspiration DI strategies, respectively. For yield per unit of total average NIW calculation, we used the average grain yield values of the irrigated seasons. These values were derived by dividing the total average NIW by the total average applied irrigation water (AIW) of the study seasons (Table 4). For instance, under full lime treatment (L1), if the yield is 8.4 Mg ha–1 at full irrigation with 510 mm of seasonal irrigation water, then the yield per total NIW under 100% ETc is indeed 8.4 Mg. However, under 80%, 60%, and 50% ETc, it is 10.4, 12.7, and 13.6 Mg, respectively, for the same 510 mm water in the 2019 irrigated season. The corresponding area under irrigation required to achieve these yields is 1 ha for full irrigation, and 1.25, 1.67, and 1.99 ha for 80%, 60%, and 50% ETc strategies, respectively. On currently non-irrigated land, crops are not cultivated due to the absence of rainfall between January and May.
Water productivity
Crop WP was calculated by dividing the yield by ETc for the control (100% ETc), or by 80%, 60%, and 50% ETc in the case of DI strategies (equation 1). The water saved due to DI and ISFM was determined by subtracting the cumulative irrigation water used under each DI treatment from the cumulative irrigation water under the full irrigation treatment (equation 2).
$$Water{\rm{\;}}productivity{\rm{\;}}\left( {WP} \right) = {\rm{\;}}{Y \over {\Sigma E{T_c}}}$$
$$Water{\rm{\;}}saved{\rm{\;}}\left( {WS} \right) = {\rm{\;}}\mathop \sum \limits_{i = 1}^n E{T_{cf}} - \mathop \sum \limits_{i = 1}^n E{T_{cd}}$$
where Y is the marketable maize grain yield (kg), and ETc is the total seasonal crop water requirement (mm) calculated by the CropWat program for the control (100% ETc) or 80%, 60% and 50% ETc in case of DI, ETcf is the irrigation water applied for full irrigation treatment, ETcd is the irrigation water applied for deficit irrigation.
Statistical analyses
Statistical analysis and graphical representation were conducted using the R Environment, version 3.4.2. (R Development Core Team, 2020). The impact of both DI and ISFM treatments on maize grain yield and biomass was assessed through two-way analysis of variance with replications. Mean separation was performed using Tukey’s Honestly Significant Difference (HSD) test in cases where the analysis of variance revealed statistically significant differences (p < 0.05) among treatments. Linear regression was employed to examine the relationship between WP and applied water amount. Prior to conducting the statistical analyses, the normal distribution of residuals and homoscedasticity were verified. The relationship between the applied water amount and WP was analyzed using linear regression.
Results
Maize grain yield per unit of land
Grain yield per ha was significantly affected by water deficit regardless of soil fertility levels (Tables 5, S1, and S4). At 80% ETc, grain yield showed reductions ranging from 1.2% to 5.1% under lime combined with manure treatments (L1 to L3), 4.1% to 5.6% under M, and 8.8% to 8.9% under C compared to 100% ETc. Under 60% ETc, grain yield reductions compared to full irrigation were 6.3% to 10.6% under L1 to L3, 13.7% to 14.1% under M, and 19.1% to 24.6% under C. Application of 50% ETc resulted in grain yield reductions of 17% to 20% under L1 to L3, 24% to 27% under M, and 25% to 28% under C.
Average maize grain yield per unit of land (Mg ha−1) under deficit irrigation (DI) and integrated soil fertility management (ISFM) strategies
Where the standard deviation of the mean value is given in parentheses. Values in a column followed by the same small letters are not significantly different (p < 0.05) between the DI treatment (ETc%) for each ISFM, while values in a row followed by the same capital letters are not significantly different between the ISFM treatment for each DI (i.e. ETc%). 100% ETc is fully irrigated and 0% ETc saved; 80% ETc is 80% ETc applied and 20% ETc saved; 60% ETc is 60% ETc applied and 40% ETc saved; 50% ETc is 50% ETc applied and 50% ETc saved. DI is deficit irrigation. ISFM is integrated into soil fertility management. L1 is a combination of 1.43 Mg ha–1 of lime with 3 Mg ha–1 of manure and full doses of urea (200 kg ha–1) plus 200 kg ha–1 of NPSB (Nitrogen, Phosphorus, Sulphur and Boron), hereafter referred to as inorganic fertilizer; L2 is a combination of 1.15 Mg ha–1 of lime with 3 Mg ha–1 of manure and full doses of inorganic fertilizer; L3 is a combination of 0.86 Mg ha–1 of lime with 3 Mg ha–1 of manure and full doses of inorganic fertilizer; M is a combination of 3 Mg ha–1 of manure along with full doses of inorganic fertilizer; and C is using only full doses of inorganic fertilizer.
Regardless of irrigation water doses, ISFM treatments significantly influenced maize grain yield per ha (Tables 5, S2, and S5). Grain yield improvement ranged from 39% to 73% under lime combined with manure treatments (L1 to L3), and 27% to 41% under M compared to C across all irrigation scenarios (100% to 50% ETc). The impact of manuring was most pronounced at 60% ETc compared to other DI strategies. Compared to M, the highest lime doses (L1) increased grain yield by 17% to 28% under all irrigation scenarios.
The interaction effect between DI and ISFM was statistically significant (Tables 5, S3, and S6), indicating that both factors can concurrently influence maize production. In both study years, the ISFM control treatment (C) differed significantly from other combined DI and ISFM treatments. In all irrigation scenarios, L1 exhibited significant differences compared to L2, L3, M, and C during the study periods. However, L2 and L3 did not show significant differences under 60% ETc across the study years. Moreover, L2 at 80% and 60% ETc was not significantly different compared to L1 at 60% ETc. When compared to L1 and L2 at 60% ETc, L3 at 60%, 80%, and 100% ETc did not show significant differences during the 2019 and 2020 irrigated seasons. In 2019, L3 and M at 100% ETc were not significantly different from L3 at 60% and 80% ETc. Specifically considering the 2019 irrigated season, L2 at 80% ETc did not exhibit significant differences from 80% and 100% ETc at L3.
The positive influence of soil fertility improvement through combined liming, manuring, and inorganic fertilizers on maize yield significantly declined with reduced water stress. However, liming and manuring notably boosted maize yield under fully irrigated conditions, with this effect lessening as water supply decreased (Table 5). Under conditions of soil fertility stress, the beneficial impact of optimal moisture availability on maize yield was notably reduced. Nevertheless, full irrigation combined with improved soil fertility management substantially increased maize yield. Yield reductions resulting from decreased water application were influenced by lime doses, with higher lime doses (L1) exhibiting relatively lower yield penalties compared to lower lime doses (L3). The most significant yield reductions were observed at C compared to treatments involving lime (L1–L3) combined with M.
Although implementing a 50% ETc irrigation regime resulted in greater water savings across all ISFM treatments, it induced noticeable stress on the crop and marginally decreased grain yield. This means that during the study period, the average yield dropped by 9%, 22%, and 26% in the ISFM control treatment at 80%, 60%, and 50% ETc.
Across all ISFM strategies, 80% ETc emerged as the optimal choice for achieving higher yields relative to the amount of water applied. However, in scenarios where water scarcity is a pressing concern and sufficient arable land is available, 60% ETc stands as the secondary option for maximizing yield with the available water resources.
Maize biomass per unit of land
Aboveground biomass exhibited significant (p < 0.05) variations influenced by DI, ISFM, and their interaction (Figure 1; Table S6). Similar to grain yield, the combined effect of DI and ISFM profoundly impacted maize production, complementing the individual significant effects of DI and ISFM. Higher biomass production was found at L1, L2, L3, and M compared to C, showing significant differences across DI application rates of 100%, 80%, 60%, and 50% ETc, as well as among ISFM practices. Additionally, under all DI levels, biomass exhibited an increasing trend with higher doses of lime combined with manuring. The ISFM control treatment (C) consistently caused lower biomass production across all irrigation scenarios during the study periods. Furthermore, biomass decreased noticeably as lime doses decreased in all irrigation scenarios. Notably, biomass found in treatments receiving lime and manure (L1–L3) under 50% ETc was comparable to those in treatments receiving only M under 100% ETc irrigation strategies. Similarly, biomass production in C under 100% ETc was lower than those in M under 50% ETc irrigation strategies. Statistically, the amount of biomass found in C under 100% and 80% ETc were comparable. At 50% ETc, L2 and L3 caused statistically similar biomass, although this trend was not observed in grain yield and WP results. Biomass decreased with increasing moisture deficit across all ISFM treatments.
Mean total biomass (two seasons average) per unit of land (Mg ha−1) under deficit irrigation and integrated soil fertility management (ISFM) strategies. Bars indicated with the same small letters are not significantly different between the ISFM under the same applied water amount (ETc, %) levels, and bars indicated with the same capital letters are not significantly different between the applied water amount (ETc, %) levels across integrated soil fertility management treatments. The vertical bars indicated the standard deviation of the mean. 100% ETc means that 100% of the crop ETc is applied, with 0% saved. For 80% ETc, 80% is applied and 20% is saved. Similarly, 60% ETc means 60% is applied and 40% is saved, while 50% ETc indicates that 50% is applied and 50% is saved. L1 is a combination of 1.43 Mg ha–1 of lime with 3 Mg ha–1 of manure and full doses of urea (200 kg ha–1) plus 200 kg ha–1 of NPSB (Nitrogen, Phosphorus, Sulphur and Boron), hereafter referred to as inorganic fertilizer; L2 is a combination of 1.15 Mg ha–1 of lime with 3 Mg ha–1 of manure and full doses of inorganic fertilizer; L3 is a combination of 0.86 Mg ha–1 of lime with 3 Mg ha–1 of manure and full doses of inorganic fertilizer; M is a combination of 3 Mg ha–1 of manure along with full doses of inorganic fertilizer; and C is using only full doses of inorganic fertilizer.
Maize yield per unit of water and water productivity
In this investigation, maize yield was evaluated relative to the total AIW as part of WP assessment (Table 6). Mean values of yield increased as the water deficit intensified across all ISFM treatments during the study periods. For instance, under L2, the average grain yield over two irrigated seasons was 8.15 Mg ha–1 under 100% ETc and 6.65 Mg ha–1 under 50% ETc. As the latter regime utilizes 50% less water, twice as much land can be irrigated with the same water amount, resulting in an overall yield of 13.3 Mg per 2 ha (Table 6). Similarly, applying 50% ETc DI under C yielded an average of 8.3 Mg per 2 ha, nearly equivalent to the overall yield at 100% ETc under L1 (8.45 Mg ha–1) and slightly higher than that under L2 (8.15 Mg ha–1) and L3 (7.85 Mg ha–1). On average, applying 50% ETc DI at M resulted in a yield of 10.7 Mg per 2 ha, representing a 49% increase compared to the overall yield at 100% ETc in the M treatment. Across all ISFM treatments, the overall yield per unit of AIW increased with reducing water application compared to full irrigation at 100% ETc, with an increasing trend from 100% ETc to 50% ETc irrigation scenarios. Within each irrigation scenario, the yield increase per unit of AIW appeared highest at L1, L2, L3, M, and C treatments, respectively, owing to the lower yield reduction attributable to liming and manuring.
Maize grain yield (Mg) per applied irrigation water amount under deficit irrigation (DI) and integrated soil fertility management (ISFM) strategies
Where 100% ETc is fully irrigated and 0% ETc saved; 80% ETc is 80% ETc applied and 20% ETc saved; 60% ETc is 60% ETc applied and 40% ETc saved; 50% ETc is 50% ETc applied and 50% ETc saved. DI is deficit irrigation. ISFM is integrated soil fertility management, and AIW is total irrigated water amount. L1 is a combination of 1.43 Mg ha–1 of lime with 3 Mg ha–1 of manure and full doses of urea (200 kg ha–1) plus 200 kg ha–1 of NPSB (Nitrogen, Phosphorus, Sulphur and Boron), hereafter referred to as inorganic fertilizer; L2 is a combination of 1.15 Mg ha–1 of lime with 3 Mg ha–1 of manure and full doses of inorganic fertilizer; L3 is a combination of 0.86 Mg ha–1 of lime with 3 Mg ha–1 of manure and full doses of inorganic fertilizer; M is a combination of 3 Mg ha–1 of manure along with full doses of inorganic fertilizer; and C is using only full doses of inorganic fertilizer.
Both DI and ISFM treatments significantly influenced crop WP throughout the study periods (Figure 2). WP was consistently highest under the greatest water-saving practices and decreased linearly with increased water application across all ISFM treatments. Particularly noteworthy is that the decrease in WP from the least to the most irrigation water saved was consistent across all lime and manure treatments (L1–L3 and M), exhibiting a similar slope of the regression line (Figure 2). Lime treatments enhanced WP by 58–67% when irrigation water was nearly halved (50% ETc) compared to full irrigation (100% ETc), while manure treatment increased WP by 49% under the same irrigation scenarios. Conversely, the C treatment displayed a relatively lower response to changes in irrigation strategy, showing a 46% increment with approximately half the amount of irrigation water compared to full irrigation.
Applied water amount (ETc, %) under varied integrated soil fertility management and average ETc water productivity of 2019 and 2020. 100% ETc means that 100% of the crop ETc is applied, with 0% saved. For 80% ETc, 80% is applied and 20% is saved. Similarly, 60% ETc means 60% is applied and 40% is saved, while 50% ETc indicates that 50% is applied and 50% is saved. Bars are the standard deviation of the mean for each treatment. L1 is a combination of 1.43 Mg ha–1 of lime with 3 Mg ha–1 of manure and full doses of urea (200 kg ha–1) plus 200 kg ha–1 of NPSB (Nitrogen, Phosphorus, Sulphur and Boron), hereafter referred to as inorganic fertilizer; L2 is a combination of 1.15 Mg ha–1 of lime with 3 Mg ha–1 of manure and full doses of inorganic fertilizer; L3 is a combination of 0.86 Mg ha–1 of lime with 3 Mg ha–1 of manure and full doses of inorganic fertilizer; M is a combination of 3 Mg ha–1 of manure along with full doses of inorganic fertilizer; and C is using only full doses of inorganic fertilizer.
Compared to C, WP increased by 19–67% under lime combined with manure treatments (L1–L3) depending on the level of DI. At M, this increment ranged between 18% and 49% across DI treatments. In all DI regimes, the highest and lowest WP was found at L1 and C compared to the other ISFM treatments, respectively. Under full irrigation (100% ETc), L1 increased WP by 6.3%, 9.7%, 18%, and 49% compared to L2, L3, M, and C, respectively. At 80% ETc, the application of L2, L3, M, and C decreased WP by 8.4%, 10.7%, 22%, and 62% compared to L1. Similarly, at 60% ETc, L2, L3, M, and C reduced WP by 9.6%, 13%, 24%, and 71% (in that order) compared to L1. Likewise, at 50% ETc, WP at L1 increased by 6.8%, 11.7%, 25%, and 62% compared to L2, L3, M, and C (in that order).
Considering DI alone, WP significantly improved with increasing levels of water deficit across all ISFM treatments. Increasing the water deficit from 100% ETc to 80%, 60%, and 50% ETc increased WP on average by 22%, 49%, and 58% at L1, respectively, over the 2019 and 2020 irrigated seasons. Similarly, WP increases ranged between 20% and 58% at L2, and between 21% and 55% at L3. At manure-treated plots applying 80%, 60%, and 50% ETc improved average WP by 18%, 41%, and 49%, respectively, with such increases varying between 13% and 46% at the control, where only inorganic fertilizers applied. This latter pattern indicates a relatively lower percentage of WP increment compared to values found at L1, L2, L3, and M. Our findings indicate that implementing 20% and 40% DI scenarios (80% and 60% ETc) under L1, L2, L3, and M would yield the best results among the tested options for attaining higher maize grain yield while conserving 20% to 40% of ETc. This saved water could then be utilized to irrigate additional land using the same amount of water as full irrigation. Therefore, the combined effect of ISFM and DI enhances WP, presenting an optimal strategy when full irrigation is unfeasible.
Effect of DI and ISFM on soil moisture
Overall, SWC levels were notably higher under treatments containing lime combined with manure (L1, L2, L3) and manure alone (M) compared to the control (C), which solely received inorganic fertilizer throughout the maize growth season (Figure 3). At 100% and 80% ETc, no water scarcity was observed during any crop growth stage under all ISFM treatments, except for treatment C. In this case, the crop experienced water shortage during the maturity stage, which extended into the ripening stage.
The average soil water content (SWC) (2019 and 2020 irrigated seasons) found one day and two days after irrigation sessions under integrated soil fertility and deficit irrigation (• indicates SWC; — indicates SWC at permanent wilting point; — indicates SWC at field capacity) for irrigated seasons (2019 and 2020) over top 100 cm soil depth. The red line indicates the water stress threshold. 100% ETc means that 100% of the crop ETc is applied, with 0% saved. For 80% ETc, 80% is applied and 20% is saved. Similarly, 60% ETc means 60% is applied and 40% is saved, while 50% ETc indicates that 50% is applied and 50% is saved. Vertical bars indicated standard deviations (±) of the mean. Init and devt stand for initial and development stages of maize growth. L1 is a combination of 1.43 Mg ha–1 of lime with 3 Mg ha–1 of manure and full doses of urea (200 kg ha–1) plus 200 kg ha–1 of NPSB (Nitrogen, Phosphorus, Sulphur and Boron), hereafter referred to as inorganic fertilizer; L2 is a combination of 1.15 Mg ha–1 of lime with 3 Mg ha–1 of manure and full doses of inorganic fertilizer; L3 is a combination of 0.86 Mg ha–1 of lime with 3 Mg ha–1 of manure and full doses of inorganic fertilizer; M is a combination of 3 Mg ha–1 of manure along with full doses of inorganic fertilizer; and C is using only full doses of inorganic fertilizer.
The reduced soil moisture availability at C could potentially be attributed to soil acidity-induced soil fertility stress. This decreased soil moisture availability during maturity and ripening stages has the potential to negatively impact maize yield. Conversely, during the initial, development, flowering, grain filling, maturity, and late stages, consistently high SWC levels were observed under lime and manure treatments (L1, L2, L3, and M) compared to treatments without lime and manure. The crop encountered mild limited soil water supply during maturity and ripening stages under L2 and L3 at 60% ETc. However, under the same DI strategy, the crop experienced greater water scarcity under M, from tasselling and to maturity and ripening stages.
Discussion
Maize grain yield per unit of land and biomass production
This study investigated the combined effects of DI and ISFM on maize production and WP for two irrigated seasons (2019 and 2020). Across all ISFM treatments, maize production per ha exhibited improvement with increasing irrigation water amounts (Table 5). However, DI treatments resulted in only slight to moderate yield reductions under all ISFM compared to full irrigation, potentially due to moisture deficit impeding maize production. These findings align with prior DI field experiments, which reported decreased maize production with increasing moisture deficit (Zhang et al., Reference Zhang, Ma, Douglas-Mankin, Han and Trout2021; Yufenga et al., Reference Yufenga, Saddique, Ajazd, Jiatuna, Khan, Mua, Azmat, Cai and Siddique2021). Similarly, earlier studies have indicated significantly lower maize yields in DI treatments compared to full irrigation (Comas et al., Reference Comas, Trout, DeJonge, Zhang and Gleason2019; Gheysari et al., Reference Gheysari, Sadeghi, Loescher, Amiri, Zareian, Majidi, Asgarinia and Payero2017) and prior research highlighted maize’s sensitivity to moisture deficit, particularly during tasselling and cob formation stages (Greaves and Wang, Reference Greaves and Wang2017; Jia et al., Reference Jia, Sun, Ali, Liu, Zhang, Ren, Zhang and Jia2016; Zhang et al., Reference Zhang, Han, Comas, DeJonge, Gleason, Trout and Ma2019).
Maize grain yields and biomass obtained from lime and manure-treated (L1, L2, L3, and M) fields surpassed those under C in all DI treatments (Table 5). The improvements in soil hydraulic properties, as well as increased soil pH and carbon content due to the amendment could explain these results in acidic Nitisols. Recent field trials confirmed that liming and manuring combined with inorganic fertilizers could reduce exchangeable acidity and raise pH levels, promoting plant root growth, water uptake, and nutrient absorption (Agegnehu et al., Reference Agegnehu, Amede, Desta, Erkossa, Legesse, Gashaw, Rooyen, Harawa, Degefu, Mekonnen and Schulz2023; Asmamaw et al., Reference Asmamaw, Janssens, Dessie, Tilahun, Adgo, Nyssen, Walraevens, Assaye, Nigate and Cornelis2023; Baumhardt et al., Reference Baumhardt, Schwartz, Howell, Evett and Colaizzi2013; Islam et al., Reference Islam, Akter, Hoque, Farzana, Uddin, Talukder, Alsanie, Gaber and Hossain2021) and then enhancing crop productivity. Islam et al. (Reference Islam, Akter, Hoque, Farzana, Uddin, Talukder, Alsanie, Gaber and Hossain2021) reported a 58.7% increase in maize yield with the application of lime, organic manure, and inorganic fertilizers over the control. The strategic use of lime combined with manure improves the availability of basic nutrients (Thakur et al., Reference Thakur, Sharma, Sankhyan and Kumar2020) and enhances nutrient use efficiency, particularly phosphorus (Islam et al., Reference Islam, Akter, Hoque, Farzana, Uddin, Talukder, Alsanie, Gaber and Hossain2021; Pan et al., Reference Pan, Baquy, Guan, Yan, Wang, Xu and Xie2020). Herein, maize grain yield was significantly higher at the highest lime dose (L1) compared to L2, L3, M, and C (in that order), consistent with prior findings (Agegnehu et al., Reference Agegnehu, Amede, Desta, Erkossa, Legesse, Gashaw, Rooyen, Harawa, Degefu, Mekonnen and Schulz2023; Opala, Reference Opala2017).
When considering DI alone, liming (L1–L3) combined with M incurred the lowest yield penalty when compared to C. The response of ISFM was most pronounced under higher water stress levels, while full irrigation showed the least effect. This suggests that liming significantly increases nutrient availability, while manuring improves soil structure, porosity, water retention, and hydraulic conductivity a topic to be further explored. Plant nutrient uptake is directly affected by SWC, as nutrients can only reach the roots through water films (Shaxson and Barber, Reference Shaxson and Barber2003). Consequently, decreasing water content associated with higher levels of DI may reduce plant nutrient uptake. However, this effect appears to be less significant in treatments receiving only manure, possibly due to enhanced physical resilience to moisture deficit resulting from increased organic matter content and microbial activity promoting soil aggregation (Thakur et al., Reference Thakur, Sharma, Sankhyan and Kumar2020). Thus, compared to the manure treatment, the control treatment exhibits a significantly greater yield penalty with increasing water stress.
To enhance maize grain yield for smallholder farmers in Koga, efficient irrigation management strategies and amelioration of acidic soil are crucial (Asmamaw et al., Reference Asmamaw, Janssens, Dessie, Tilahun, Adgo, Nyssen, Walraevens, Assaye, Nigate and Cornelis2023; Asmamaw et al., Reference Asmamaw, Janssens, Dessie, Tilahun, Adgo, Nyssen, Walraevens, Pue De, Yenehun, Nigate, Sewale and Cornelis2022). Despite the lowest grain yield obtained from C, it still surpassed the national average maize grain yields (3.4 Mg ha–1) and lagged behind the global average (5.6 Mg ha–1) (CSA, 2021). The lowest yield found at C under 50% ETc may be partly attributed to aluminum toxicity, limiting root growth and impairing plant water absorption, as well as nutrient inaccessibility. Such lowest grain yield may also be partly attributed to moisture stress at C under 50% ETc. Conversely, the highest grain yields were found under the same amount of water supplied at L1, L2, and L3. As previously explained, increased yield can be attributed to liming and manuring, which improve soil quality, particularly soil porosity, water retention, and microbial activity, thus enhancing water and nutrient accessibility to plants. When full irrigation is not feasible, adopting 80% ETc for biomass production could be advantageous as the biomass reduction due to water scarcity is relatively minor compared to the benefits gained from irrigating additional land using the conserved water. However, in cases of severe water scarcity, farmers may opt for 60% ETc while taking into account the prevailing soil fertility conditions.
Maize grain yield per unit of water and water productivity
The overall maize grain yield per unit of AIW was enhanced across all ISFM treatments (Table 6). This suggests the potential for expanding the current irrigated area by utilizing the saved water to irrigate additional land. In comparison to full irrigation, adopting 80%, 60%, and 50% ETc irrigation strategies typically enables the irrigation of 23%, 67%, and 99% more land, respectively, with the same amount of water. This signifies a viable approach for mitigating water scarcity in regions like Koga, where moisture is a limiting factor for crop production. Despite the Koga irrigation scheme being designed for over 7004 ha of land, inefficient water use has resulted in only 5000 ha being irrigated during the first round of the irrigation season. Thus, many farmers initially included in the project are now excluded from the irrigation program. Additionally, numerous farmers residing near the Koga irrigation project express a keen interest in irrigating their land and have voiced complaints to the government about their exclusion from access to irrigation water (Asmamaw et al., Reference Asmamaw, Janssens, Dessie, Tilahun, Adgo, Nyssen, Walraevens, Assaye, Nigate and Cornelis2023). Implementing irrigation scenarios based on our findings could alleviate these issues by enabling the irrigation of a larger area within the scheme, including lands near the Koga irrigation project.
Our results indicate that soil amendment with lime and manure represents a promising strategy to maximize the benefits of DI, as soil fertility management significantly mitigated the yield penalty associated with DI. Similar findings were reported by Zahra et al. (Reference Zahra, Aftab and Haide2021) in a study on DI combined with biochar and compost, highlighting the potential of soil amendment to enhance crop yield even under water stress conditions by improving water use efficiency. In fact, water scarcity is increasingly limiting viable agricultural production, and enhancing WP could help address this challenge. The combined use of water-saving strategies like DI and ISFM strategies has the potential to increase WP, with DI reducing water use and ISFM boosting crop yield. Across all irrigation treatments, the impact of liming and manuring on WP followed the order of L1 > L2 > L3 > M > and C (Figure 2). WP at 50% ETc was notably 49–67% higher in lime and manure treatments compared to the control conditions. These findings are consistent with recent field research on the effects of DI and ISFM on wheat production and WP (Asmamaw et al., Reference Asmamaw, Janssens, Dessie, Tilahun, Adgo, Nyssen, Walraevens, Assaye, Nigate and Cornelis2023). They also align with a recent meta-analysis study by Asmamaw et al. (Reference Asmamaw, Janssens, Dessie, Tilahun, Adgo, Nyssen, Walraevens, Fentie and Cornelis2021a), which reported a 109% increase in WP by saving 65% of ETc while only reducing maize yield by 27%. The enhanced WP can be attributed to improvements in soil fertility conditions and the implementation of water-saving technologies with minimal impact on yield (Baumhardt et al., Reference Baumhardt, Schwartz, Howell, Evett and Colaizzi2013; Yufenga et al., Reference Yufenga, Saddique, Ajazd, Jiatuna, Khan, Mua, Azmat, Cai and Siddique2021).
Soil water content
The consistently higher SWC observed under 100% and 80% ETc across all growth stages in lime and manure treatments (L1, L2, L3, and M), compared to treatment C, could possibly be attributed to improvements in soil fertility resulting from liming and manuring the acidic soils (Figure 3). However, when the DI scenario was reduced to 60% ETc, the crop experienced moderate water scarcity during the maturity and ripening stages under treatments L2 and L3. Similarly, the crop experienced an even greater water deficit at 60% ETc under treatment M. Specifically, the scarcity began at the tasseling stage and continued through maturity and ripening. This could be attributed to soil acidity, which reduces the soil’s water-holding capacity and similar finding was reported by Asmamaw et al. (Reference Asmamaw, Janssens, Dessie, Tilahun, Adgo, Nyssen, Walraevens, Pue De, Yenehun, Nigate, Sewale and Cornelis2022) in their study on the effects of DI and ISFM on wheat cultivation.
Under the same DI scenario, treatment C caused higher water scarcity across all growth stages, potentially affecting stomatal closure and photosynthesis (Asmamaw et al., Reference Asmamaw, Janssens, Dessie, Tilahun, Adgo, Nyssen, Walraevens, Assaye, Nigate and Cornelis2023). At 50% ETc, significant soil moisture scarcity was observed under all lime and manure treatments (L1, L2, L3), likely leading to stomatal closure during the development, flowering, maturity, and ripening stages. Comparatively, at 50% ETc, water scarcity was more severe under treatment M than under L1, L2, or L3, beginning during the initial and development stages and extending through flowering, maturity, and late stages, intensifying as the crop advanced through its growth stages. Under the same DI level applied to treatment C, the crop encountered even more pronounced water scarcity across all growth stages. These findings suggest that soil fertility stress, possibly due to soil acidity, exacerbated the water scarcity experienced under C. Consistent with these results, a recent field study by Asmamaw et al. (Reference Asmamaw, Janssens, Dessie, Tilahun, Adgo, Nyssen, Walraevens, Assaye, Nigate and Cornelis2023) on the effects of DI and ISFM on wheat production and WP also found that the crop experienced more severe water scarcity under treatment C compared to treatments L1, L2, L3, and M at all growth stages.
Conclusion and recommendations
We investigated the combined effects of deficit irrigation (DI) and integrated soil fertility management (ISFM) on maize production and water productivity (WP) for two irrigated seasons. Our findings support the notion that implementing enhanced irrigation practices combined with ISFM can enhance crop WP. Specifically, utilizing a full dose of lime combined with manure markedly increased maize grain yield, biomass, and WP across all irrigation scenarios compared to applying 80% or 60% of the lime requirement (LR) combined with manure. Moreover, compared to a control treatment that received no lime or manure (C), the application of manure also significantly boosted maize grain yield, biomass, and WP under varying irrigation conditions. This improvement can be attributed to the increase in organic matter and the enhancement of essential nutrients available to plants across different irrigation levels. Our study also highlights the potential for expanding the current irrigated area by utilizing the saved water to irrigate additional land. This presents an opportunity for farmers who currently face water shortages and are unable to irrigate their land to benefit from this irrigation scheme. To enhance WP with minimal grain yield reduction, maize could be irrigated at 80% ETc across all ISFM strategies. For greater benefits from DI, investing in lime combined with manure appears to be a promising strategy for farmers in the Koga and other similar agro-ecologies and soil conditions.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S0014479725000092
Acknowledgements
We express our sincere gratitude to VLIR-UOS for their funding support through the Institutional University Cooperation with Bahir Dar University, Ethiopia. Our heartfelt appreciation goes out to the farmers who participated in the field experiments, whose contributions were invaluable to the success of this study. We extend our thanks to Bahir Dar University for generously providing 2 hectares of land for our experiments conducted during the study periods. Additionally, we are grateful to the Ethiopian National Meteorological Service Agency for providing us with the meteorological information essential for our research. We also acknowledge the support received from the VLIR-UOS project under the Global Minds Fund for Short Research Stays and Short Initiative project (SI).
Open access
