Soil fertility and water scarcity problem has driven more innovative approaches in agriculture to strengthen subsistence and food security especially among agricultural small holder farmers through adoption of the suitable technologies (Nyang’au et al., 2021). To curb the problem of water scarcity and low crop yields, soil and water conservation strategies, irrigation, planting trees and improved crop seeds have been used as remedies of climate change (Gebru et al., 2020; Wawire et al., 2021). The small holder farmers in the drylands of Kenya have used various strategies to improve soil fertility, soil water availability and overall crop yield hence improving the economic returns (Mati, 2006; Kimaru-Muchai et al., 2020). This involves the use of soil and water conservation practices to improve livelihood and reduce vulnerability to drought (Sawadogo, 2011; Funk et al., 2020). Water harvesting technologies such as pitting is a form of water conservation strategy used to improve water supply for agricultural use and reduce vulnerability in an event of prolonged drought (Liang and Van Dijk, 2011; Patle et al., 2020). Pitting increases the moisture content in the soil and restores productivity in areas where rainfall is insufficient (Biazin et al., 2012; Kimaru-Muchai et al., 2020; Ndeke et al., 2021).

Zai pit a form of soil and water conservation technique harvests rainwater in small pits with manure buried in the pits to improve soil fertility as well as conserve soil moisture (Danjuma and Mohammed, 2015; Danso-Abbeam et al., 2020). The pits are best suited in areas that have low annual rainfall ranging from 300 to 800 mm (Motis and Lingbeek, 2013; Mwangi, 2020). In Africa, the zai pit system has been used in Mali, Niger and Burkina Faso with studies indicating that they improve soil fertility over time (Zougmoré et al., 2003; Danso-Abbeam et al., 2020). In Kenya, the most common type is the "five by nine" pit system with the zai pit system being promoted in the arid and semi-arid lands  (ASALs)  (Mati,  2006).  The zai pit system has been practiced in Eastern Kenya which has been proved successful (Kathuli and Itabari, 2014; Kimaru-Muchai et al., 2020).

Declining soil fertility also remains a major biophysical challenge in crop production among the smallholder farmers in the drylands of Eastern Kenya. This is a result of high rates of soil erosion, continuous cultivation without adequate addition of fertility inputs and the removal of crop residues from the field (Njeru et al., 2011; Aniah et al., 2019; Sarkar et al., 2020). Integrated soil fertility management (ISFM) technologies are used to enhance agricultural productivity through maximizing and efficient use of the applied nutrients (Vanlauwe et al., 2010; Bayu, 2020). The ISFM technologies include; use of organic matter, inorganic fertilizers, use of improved germplasm and adaptation to the local conditions to improve the soil properties and the overall crop yield (Ejigu et al., 2021; Mucheru-Muna et al., 2021). The small holder farmers in the drylands of Kenya commonly use crop residues, mineral fertilizers, farmyard manure and composts for short-term supply of nutrients with most farmers believing that the use of inorganic fertilizers is the quickest and surest method to supply nutrients (Mucheru-Muna et al., 2021).

Challenges facing small holder farmers arid and semi-arid lands (ASALs) of Kenya include soil fertility decline, water scarcity, low crop yields and low economic returns. To address the challenges, a trial was set up in Kabati, Kitui County with the aim of assessing the effects of zai pits combined with ISFM technologies on soil nutrients, sorghum yields and the economic feasibility of using zai pits combined with integrated soil fertility management technologies on sorghum production within the study area.

Description of the study area  

The experiment was conducted in Kabati, Kitui County Kenya (1°14'13.0"S 37°54'52.2"E) (Figure 1). The area experiences unreliable rainfall ranging from 250 to 1050 mm per annum. The rainfall pattern is bimodal with the short rains (SR) occurring between October and December and the long rains (LR) being experienced between March and May. The rest of the year is dry. The hot months are July-September and January-February with high temperatures ranging between 14 and 34°C.

The topography of the area is described as hilly rugged uplands and lowlands. The general landscape of the area is flat with plains gently rolling down towards the east where the altitudes are as low as 400 m above sea level.

The soils found at Kabati are lixisols consisting of strongly weathered, leached and finely textured materials with high base saturation (Nezomba, 2016). As described by IUSS WG WRB (2015), a publicly available 30-m Soil Information System of Africa, lixisols are classified as oxic subgroups of Alfisols with principal qualifiers including, gleyic, stagnic, ferralic, nudiargic, lamellic, albic, ferric, rhodic/chromic, xanthic, gypsic, calcic, fractic, skeletic and haplic. The supplementary qualifiers include abruptic, albic, ferric, geric, humic, hyperochric, lamellic, profondic, and stagnic. The soils are formed due to chemical weathering and are found in subtropical to warm climate with a pronounced dry season. They are thin and brown with low levels of available nutrients and low organic carbon levels. The surface soils have low aggregate stability prone to erosion and slaking when exposed to raindrops. Tillage and erosion measures to control erosion include mulching, use of cover crops and terracing to conserve them.

Kabati is in the upper midland 3-4 (semi-arid farming zone) with various natural resources including forests, rivers, hills and wildlife. The semi-arid zone has good agricultural potential used for agricultural production but frequent crop failures are recorded. The main socio-economic practices in the area include crop, livestock production, fish production and tourism. The main food crops grown in the area include legumes such  as  green  grams (Vigna  radiata), cowpeas (Vigna unguiculata), pigeon peas (Cajanus cajan) and beans (Phaseolus vulgaris). Cereals include maize (Zea mays), sorghum (Sorghum bicolor) and millet (Pennisetum glaucum) with tuber crops being cassava (Manihot esculenta) and sweet potatoes (Ipomoea batatas). Fruits include avocado (Persea americana), watermelon (Citrullus lanatus), pawpaw (Carica papaya) and mangoes (Mangifera indica).

Rainfall trend over the experimental period

The two growing seasons (SR2018 and LR2019) received varying amounts of rainfall with the total seasonal amount of rainfall recorded being 382.3 and 116.3 mm, respectively (Figure 2).  

SR2018 season (October-December, 2018) received more rainfall compared to LR2019 (March-May, 2019) which experienced less rainfall. The rains in the SR2018 growing season were uniformly distributed with the LR2019 experiencing rain in only 10 days of the growing period with the rest of the season being dry. 

Experimental layout and management

The experiment was laid out as a randomized complete block design (RCBD) with eight (8) treatments replicated thrice (Table 1). The plot size measured 6 by 4.5 m, a spacing of 1 m between the plots and 1 m for the guard zone.  The zai pits were made by digging out the topsoil to make 60 cm wide, 60 cm long and a depth of 30 cm (Danso-Abbeam et al., 2020). The topsoil was then refilled to a level of 15 cm before the fertility amendments were added. Sorghum gadam variety a drought-tolerant crop was planted at a spacing of 75 and 20 cm inter and intra-row, respectively and two sorghum plants were planted per hill. Weeding was done twice during the planting period by use of a hand hoe. No diseases were observed on the sorghum during the experimental period. Six out of the eight treatments, had three external fertility amendments inputs in both zai and conventional planting (1) Cattle manure (60 kg N ha^-1), (2) Cattle manure (30 kg N ha^-1) + 30 kg N ha^-1 / 30 kg P ha^-1,  and (3) 60 kg N ha^-1 / 60 kg P ha^-1 were applied at the beginning of every season to give an equivalent of 60 kg N ha^-1 and 60 kg P ha^-1 the KARLO recommended rate of N and P for sorghum totalling to six treatments (Karanja et al., 2014). The seventh and eight treatments were absolute control in both conventional and zai planting where no external fertility inputs were applied (Table 1).

Data collection

Soil sampling and laboratory analysis

Soil sampling was done before setting up the experiments at the field in the SR2018 season and at the end of the LR2019 season in all the plots at a depth of 0 to 15 cm by use of a soil auger. The soil samples were collected from five different spots within the plot then bulked to one sample which was then analysed for total nitrogen (TN), soil pH, phosphorus, electrical conductivity, and organic carbon. 

Soil laboratory analysis

Soil chemical properties

The soil samples from the field were delivered to the laboratory, air-dried and ground to pass through a 2 mm sieve.  All the laboratory analyses on the chemical properties of the soil were performed using the standard methods for analysing soils as described by Motsara and Roy (2008). Total nitrogen was analysed using the Kjeldahl method as described by Bremner (1960), available phosphorus (P) was measured using the Brays method, potassium (K) was estimated with a flame photometer, soil pH was measured by an electronic pH meter, the electric conductivity of the soil was measured by electric conductivity meter, and organic carbon (OC) was measured by the ignition method (Motsara and Roy, 2008).

Soil physical properties

Soil aggregate stability (dry sieving) was determined as described by Ekwuea et al. (2018), soil particle size was determined using the hydrometer method and soil bulk density was determined by core sampling method described by Motsara and Roy (2008). Table 2 shows the averages for the initial soil chemical and physical properties at Kabati, Kitui County.

Sorghum yields

Sorghum heads were harvested at maturity from the net area of each plot and weighed. The sorghum heads were then sundried, threshed and the dry weight recorded as  well. Grain  moisture  was determined using a moisture meter and the grain weight was adjusted to 12.5% moisture content after sun drying.

Rainfall measurement

Rainfall data was collected using a simple rain gauge calibrated in millimetres. Daily rainfall was measured and recorded with monitoring visits done to ensure the accuracy of the rainfall.

Economic analysis

To evaluate the economic returns of using zai pits and integrated soil fertility management (ISFM) technologies on sorghum production, partial budgeting was used to compare the costs and financial benefits of each treatment. The benefits and costs with each treatment were evaluated using the different ISFM technologies with zai pits. The costs included fertility-enhancing inputs (manure and inorganic fertilizer), pitting and labour with benefits being increased sorghum yield. The inputs and outputs prices derived from the prevailing market prices in the area and values used in economic analysis are shown in Table 3.

Detailed data on the labour requirements were collected in the two seasons for all the various field operations (land preparation, manure and fertilizer application, weeding, harvesting and pitting) and time taken for each activity was recorded using a stopwatch. Labour was valued at the local wage rate (USD man day^-1; USD 3.532) per working day (8 h). Net benefit, return to labour and benefit-cost ratio were used as the economic tools in assessing profitability. The net benefit, return to labour and benefit-cost ratio were calculated using the following formulas:

Net benefits=Total benefits-Total costs

Return to labour=Net benefits/Labour costs

Benefit cost ratio=Net benefits/Total Cost

Data analysis

Data on sorghum yield, soil chemical properties and economic return were subjected to analysis of variance (ANOVA) using the proc procedure in SAS 9.2 software and means separated using standard error of differences (SED) at p<0.05. To compare the effects of each treatment on sorghum yield, conversion of relative increases compared to the control were done. Changes in yield were compared between the two seasons using t-test. Comparisons on soil  nutrients  from  soil  samples  collected at the beginning and the end of the experiments were analysed using t-test.

Interaction effects of pits, cattle manure and mineral fertilizers on selected soil chemical properties

The treatments significantly (p=0.0214) reduced total nitrogen at the end of the experiment with electrical conductivity and available phosphorous increasing significantly (p=0.0097, p=0.0005, respectively) over the same period. Under the zai tillage system nitrogen content reduced significantly (p=0.0390, p=0.0310, p=0.0005, respectively) in zai pit with manure and half rate mineral fertilizer, zai with sole manure and zai without input treatments (Table 4). Under the conventional tillage system nitrogen content decreased significantly (p=0.0212, p=0.0198, p=0.0007, respectively) in conventional with manure and fertilizer, conventional with sole manure and conventional with mineral fertilizer treatments. Zai with fertilizer had the highest percentage reduction (-65.22%) and conventional with manure recording the lowest percentage reduction (-38.71%).

Under the conventional tillage system, electrical conductivity significantly (p=0.0037, p=0.0415, p=0.0180, respectively) increased in conventional with fertilizer, conventional with manure and fertilizer and conventional without input. Under the zai tillage system, electrical conductivity significantly (p=0.0010, p=0.0175, p=0067, respectively) increased in zai with fertilizer, zai with sole manure and zai with manure and mineral fertilizer (Table 4). Soil organic carbon (SOC) significantly (p=0.0015, p=0.0109, p=0.0146, respectively) reduced in conventional without inputs, zai with manure and mineral fertilizer and zai with mineral fertilizer treatments. Available phosphorous significantly (p=0.0258, p=0.0114, p=0.0266, respectively) increased in zai planting with sole manure, zai planting with full-rate fertilizer and conventional planting with sole manure treatments (Table 4).

Zai pits, cattle manure and mineral fertilizers interaction effects on sorghum yields

Variation of grain and stover yields due to seasons, tillage systems (zai and conventional) and the different fertility inputs were observed over the study period. During the SR2018 season, sorghum grain and stover yields were significantly (p<0.0001, p=0.0005, respectively) influenced by zai pits and ISFM technologies (Table 5). During the LR2019 season the stover yields were significantly (p=0.0007) influenced by zai pits and ISFM technologies.

The highest grain and stover yields (4.37 and 12.71 t ha^-1), respectively were recorded during the short rains (SR2018) season while the lowest grain (0.06 t ha^-1) and stover (1.29 t ha^-1) yields were recorded during the long rains (LR2019) season (Table 5). The average grain and stover yields were higher under zai planting compared to conventional planting in both the SR2018 and LR2019 seasons. During the SR2018 season, grain yield was highest in zai planting with manure and fertilizer (4.37 t ha^-1) and lowest in conventional without inputs (2.06 t ha^-1) treatments. During the LR2019, the highest grain yield (Table 5) was recorded in zai planting with fertilizer (0.28 t ha^-1) and the lowest grain yield recorded in conventional with manure and mineral fertilizer treatment (0.06 t ha^-1).

During the SR2018 season, significant higher grain yields were recorded in zai with mineral fertilizer, zai planting with sole manure and zai with manure and fertilizer (p=0.0027, p=0.0036, p=0.0116, respectively) compared to conventional with fertilizer, conventional with sole manure and conventional with manure and mineral fertilizer. During the same season, stover yields were significantly higher (p=0.0009) in zai planting with manure and mineral fertilizer as compared to conventional planting with manure and mineral fertilizer. During the LR2019 season, significantly higher stover yields (p=0.0498, p=0.0390, respectively) were recorded in zai planting with mineral fertilizer and zai with sole manure as compared to conventional with mineral fertilizer and conventional with sole manure (Table 5).

Economic returns

During the SR2018 the highest net benefit was recorded in zai planting with manure and mineral fertilizer (2690.51 USD ha^-1) and the lowest recorded in zai planting without input (1228.99 USD ha^-1) in the same season (Figure 2). In the LR2019 season the net benefits recorded a negative change  (Table  6)  with  the  highest  net benefit recorded in zai with fertilizer (27.11 USD ha^-1) and the lowest recorded in conventional with manure and fertilizer (-199.71 USD ha^-1). During the SR2018 season, the highest return to  labour  was recorded in zai with sole manure treatment (22.60 USD ha^-1) and the lowest recorded in conventional with fertilizer treatment (7.41 USD ha^-1). In the LR2019 season the highest return  to  labour  was recorded in zai with fertilizer treatment (0.19 USD ha^-1) and the lowest in conventional with manure and fertilizer (-0.96 USD ha^-1). BCR was highest in zai with sole manure  (17.09 USD ha^-1) and lowest in conventional with fertilizer (6.34 USD ha^-1) during the SR2018 season. During the LR2019 season, BCR was highest (Table 6) in zai with fertilizer (0.15 USD ha^-1) and lowest in conventional with manure and fertilizer (-0.83 USD ha^-1).

The three economic tools, that is, the net benefits, return to labour and benefit cost ratio (BCR) were significantly (p=0.0048, p=0.0001, p=0.0039, respectively) influenced by zai pits and ISFM technologies during the SR2018 season. During the SR2018 season, return to labour was significantly higher (p=0.0269, p=0.0252, p=0.0379, respectively) in zai with fertilizer, zai with manure and zai with manure and fertilizer as compared to conventional with fertilizer, conventional with manure and conventional with manure and fertilizer (Table 6).

The   incorporation   of   fertility   inputs  negatively influenced total nitrogen with electrical conductivity and available phosphorous positively influenced by the end of LR2019 cropping season. The addition of manure and mineral fertilizer significantly reduced total nitrogen by the end of the two cropping seasons and this could be because the application of fertility inputs makes nitrogen readily available for plant uptake hence the reduction of total nitrogen in the soil. The reduction of total nitrogen could also be attributed to loss through volatilization, nitrous oxide emission, leaching, erosion and oxidation of nitrogen as reported by Pal et al. (2020). These results corroborated by Pasley et al. (2019) who noted that an increase in nitrogen fertilizer application increased nitrogen uptake by crops reducing its availability in the soil. Similarly, Omara et al. (2019) reported that the efficiency of nitrogen uptake by crops was accelerated by manure and fertilizer application. However, other studies recorded a significant increase in total nitrogen  in  the soil. For  instance,  Bedada  et  al. (2014) noted that a combination of compost and nitrogen, phosphorous (NP) fertilizer increased the stock of total nitrogen 0-10 cm compared to the control experiment and nitrogen phosphorous (NP) fertilizer alone treatments. Similarly, Mattuso et al. (2014), Yegon et al. (2016), Kihara et al. (2016) and Liu et al. (2020) reported that the addition of manure and mineral fertilizer increased total nitrogen in the soil.

The application of sole manure increased the electrical conductivity of soil. This could be attributed to the high amounts of dissolved salts in manure. Soil electrical conductivity increases as a result of manure application is related to the high amounts of dissolved salts that is beneficial in supplying a pool of nutrients and ions into the soil.  Similar findings have also been reported by Carmo et al. (2016a, b) and Miller et al. (2016) who indicated that electrical conductivity increased with manure application.

Conversely, soil organic carbon reduced in all the  treatments  by  the  end  of  the  two  cropping seasons. This could be attributed to the slow changes in soil organic carbon in compacted, poorly drained soils and clayey soils compared to sloping and coarse-textured soils. Tillage also speeds up the loss of soil organic carbon by increasing its mineralization and loss by erosion. This is because mixing of soils with litter favours bacteria and promotes the rapid breakdown process of organic carbon. Tillage induced erosion of soils is also the cause of severe loss of soil organic carbon more especially in upland landscapes. Similar results have also been reported by Liu et al. (2003), Blanco-Canqui et al. (2013) and Corsi et al. (2012) who indicated that soil organic carbon declined significantly in the first five years of cultivation. Long-term experiments are required to detect changes in soil organic carbon because it responds slowly to changes in agricultural management. Therefore, this means that the changes may require a long period of time to be detected and quantify the effect of the management activities (Haddaway et al., 2015). A similar study on the changes of soil organic carbon in an established plantation in northern China recorded an initial decrease in soil carbon (Lei et al., 2019). This could be attributed to the loss of carbon through decomposition which outweighed gains of carbon from litter.

Available phosphorous increased in treatments that had sole manure in zai and conventional tillage systems an indication that manure could have contributed to the increase in phosphorous in the soil. The current trend of results is in consonance with findings of Ali et al. (2019) who noted that manure application as an amendment in agricultural soils improved the soil physiochemical properties and cycling of nutrients through enhancing enzyme and soil microbial activities leading to improved phosphorous bioavailability for crop uptake. Buckley and Makortoff (2004) reported that manure contains about 45 to 90% of inorganic orthophosphates a form in which phosphorous is taken up by plants which makes it a rich supplier of phosphorous into the soil. 

On average, the LR2019 season recorded lower grain and stover yields compared to the SR2018 season. This could be associated with the dry spells and prolonged meteorological drought that was experienced in the planting (LR2019) season where the cumulative rainfall declined (Figure 2) affecting the growth and production of sorghum. Similar results were also reported by Rockström (2010), Ibrahim et al. (2011) and Nyakudya and Stroosnijder (2011) who noted that low yields experienced were as a result of water stress which is often experienced by the smallholder farmers in the arid and semi-arid regions.

Zai pits as a water harvesting technique increased the grain yields in SR2018 and this could be attributed to their ability to retain soil moisture and improve nutrient efficiency. These results were similar with Amede et al. (2011) and Wouterse (2017) who noted that the zai pit technology was an intervention used by smallholder farmers   to   increase   agricultural   productivity   through improving precipitation capture, reduction of runoff, increase water infiltration and water evaporation from the soil. The current trend of results also corroborated by Kathuli and Itabari (2014) who reported that the use of zai pits significantly increased sorghum grain yield.  Similarly, Mazvimavi and Twomlow (2008) noted higher yields in pitting technology compared to the conventional tillage system.

During the SR2018 season, grain yields were significantly higher in zai with full-rate fertilizer compared with zai without input an indication that mineral fertilizer significantly increased the grain yields. This could be because mineral fertilizer improves soil chemical and physical properties by increasing the availability of nutrients in the soil and promoting the growth of crops.

During the same season grain yields were significantly higher in zai with sole manure and zai with manure and mineral fertilizer compared to the control. This was also attributed to the ability of the combination to enhance release of nutrients and uptake by crops. Tittonel et al. (2008), Kihara et al. (2017) and Mi et al. (2018) attributed the increase of grain yield to the use of fertilizer and manure. Matusso et al. (2014) linked the significant increase in grain yield to the ability of the combination to enhance the nutrient release and uptake. Amede et al. (2011) also reported that zai pits and a combination of fertilizer additions increased the yield of potatoes by 500 to 2000% and bean yield by 250%. The application of manure alone or in combination with mineral fertilizer increased crop yield and this could be associated with the increase in the supply of the nutrients in the soil as well as the ability of the combination to enhance release of nutrients hence increase in nutrient availability. Chen et al. (2018) reported that the application of organics alone or a combination with mineral fertilizers led to increased crop yield compared to the sole mineral fertilizers. Elsewhere, Chivenge et al. (2011) reported an increase in maize yield in treatments that had a combination of organic resources and fertilizers (114%) and sole organic resources (60%). Biazin et al. (2012), Dunjana et al. (2012) and Kar et al. (2013) also noted that rainwater harvesting in combination with the use of both inorganic and organic inputs increases the nutrients in the soil improving crop productivity.

The significant difference in the economic parameters between the seasons, with SR2018 performing much better than the LR2019 season could be as a result of the low grain yields recorded in LR2019 due to the prolonged dry spell. Ray et al. (2018) noted that drought causes significant reduction in yields and economic returns in both irrigated and rainfed crops because of the reduction of water and moisture available for crop growth. The SR2018 season had higher benefits in treatments that had a combination of manure and fertilizer in both conventional and zai tillage systems compared to the control. This was an indication that the combined use of organics and mineral fertilizers  was  economically  viable

to the small-holder farmers. Hobbs et al. (2011) and Kebede (2020) attributed the high net benefits with soil fertility amendments and water conservation techniques used in crop production. This generally explains the higher net benefit results in the SR2018 season that on average the combination of cattle manure and mineral fertilizers had higher net benefits than the recommended rate of fertilizer (Olarinde et al., 2012; Girma et al., 2020).

The net benefits, return to labour and BCR were higher in zai treatments compared to conventional treatments with similar fertility inputs. This could be attributed to the water conservation and fertility technologies in the zai treatments which increased the overall net benefits.  Hobbs et al., (2011) reported that zai pits has been used as an intervention to increase productivity by improving water availability and nutrient efficiency. Higher net benefits, return to labour and BCR were recorded in treatments that had a combination of manure and mineral fertilizer in both zai and conventional planting. This could be attributed to the increased supply of nutrients hence high productivity. Similar results have been reported by Mutegi et al. (2012) that on average, the combined use of organic inputs and mineral fertilizers had higher net benefit and BCR compared to the sole application of mineral fertilizers. Kearney et al. (2012), Ojiem et al. (2014), Matusso et al. (2014) and Thimmaiah et al. (2016) noted that greater net benefits were recorded in a combined application of inorganics and mineral fertilizers compared with the application of sole inorganics and sole mineral fertilizer.

During the SR2018 season, BCR was greater than one in all the treatments compared with the LR2019 season whereby the BCR was less than one in all the treatments. If the BCR is greater than one, it shows that the technologies can be beneficial as Shively and Galopin (2013) reported. Higher BCR was recorded in zai planting with manure and zai with manure and mineral fertilizer compared to their conventional counterparts in the two cropping seasons. This could be a consideration because it is more economical and a feasible alternative available for nutrient supplementation compared to the high costs of purchasing inorganic fertilizers (Mucheru-Muna et al., 2007). The combined use of fertilizer and inorganics can be practiced by small-holder farmers in the study area to supplement nutrient deficiencies due to the limited purchasing power of the farmers.

Cattle manure an amendment used to improve soil fertility increased soil pH in both zai pit and conventional planting. The application of cattle manure and mineral fertilizer improved the soil electrical conductivity and available phosphorous. Total nitrogen and organic carbon significantly reduced at the end of the LR2019 cropping season due to losses. There were also beneficial effects in zai planting as compared to conventional  planting  and combined application of manure and mineral fertilizer relative to applying sole fertilizer in terms of sorghum grain yields and economic benefits. Grain and stover yields were higher in zai planting compared to their conventional counterparts with the same fertility inputs. Higher net benefits, return to labour and BCR were also observed in treatments that had a combination of manure and mineral fertilizer. Rainfall variability in amount and distribution greatly affected sorghum yields across the seasons. The installation of zai pits is more labour intensive but the economic returns are higher in subsequent cropping systems compared to the conventional tillage system. Therefore, there is a need to integrate organic and inorganic fertilizer as a viable option and use of zai pits by the small holder farmers to enhance crop production and profitability.

The authors have not declared any conflict of interests.