Agriculture is essential to a flourishing economy. Although soil is essential for sustainable food production, its quality can decline as cultivation becomes more intensive and demand increases. The importance of healthy soil cannot be overstated, as a lack of nutrients can significantly lower crop yield. Smart soil prediction and digital soil mapping offer accurate data on soil nutrient distribution needed for precision agriculture. Machine learning techniques are now driving intelligent soil prediction systems. This article provides a comprehensive analysis of the use of machine learning in predicting soil qualities. The components and qualities of soil, the prediction of soil parameters, the existing soil dataset, the soil map, the effect of soil nutrients on crop growth, as well as the soil information system, are the key subjects under inquiry. Smart agriculture, as exemplified by this study, can improve food quality and productivity.

Cassava-maize intercropping is a common practice among smallholder farmers in Southern Nigeria. It provides food security and early access to income from the maize component. However, yields of both crops are commonly low in farmers’ fields. Multi-locational trials were conducted in Southern Nigeria in 2016 and 2017 to investigate options to increase productivity and profitability through increased cassava and maize plant densities and fertilizer application. Trials with 4 and 6 treatments in 2016 and 2017, respectively were established on 126 farmers’ fields over two seasons with a set of different designs, including combinations of two levels of crop density and three levels of fertilizer rates. The maize crop was tested at low density (LM) with 20,000 plants ha⁻¹ versus high density (HM) with 40,000 plants ha⁻¹. For cassava, low density (LC) had had 10,000 plants ha⁻¹ versus the high density (HC) with 12,500 plants ha⁻¹.; The fertilizer application followed a regime favouring either the maize crop (FM: 90 kg N, 20 kg P and 37 kg K ha⁻¹) or the cassava crop (FC: 75 kg N, 20 kg P and 90 kg K ha⁻¹), next to control without fertilizer application (F0). Higher maize density (HM) increased marketable maize cob yield by 14 % (3700 cobs ha⁻¹) in the first cycle and by 8% (2100 cobs ha⁻¹) in the second cycle, relative to the LM treatment. Across both cropping cycles, fertilizer application increased cob yield by 15 % (5000 cobs ha⁻¹) and 19 % (6700 cobs ha⁻¹) in the FC and FM regime, respectively. Cassava storage root yield increased by 16 % (4 Mg ha⁻¹) due to increased cassava plant density, and by 14 % (4 Mg ha⁻¹) due to fertilizer application (i.e., with both fertilizer regimes) but only in the first cropping cycle. In the second cycle, increased maize plant density (HM) reduced cassava storage root yield by 7% (1.5 Mg ha⁻¹) relative to the LM treatment. However, the negative effect of high maize density on storage root yield was counteracted by fertilizer application. Fresh storage root yield increased by 8% (2 Mg ha⁻¹) in both fertilizer regimes compared to the control without fertilizer application. Responses to fertilizer by cassava and maize varied between fields. Positive responses tended to decline with increasing yields in the control treatment. The average value-to-cost ratio (VCR) of fertilizer use for the FM regime was 3.6 and higher than for the FC regime (VCR = 1.6), resulting from higher maize yields when FM than when FC was applied. Revenue generated by maize constituted 84–91% of the total revenue of the cropping system. The highest profits were achieved with the FM regime when both cassava and maize were grown at high density. However, fertilizer application was not always advisable as 34 % of farmers did not realize a profit. For higher yields and profitability, fertilizer recommendations should be targeted to responsive fields based on soil fertility knowledge.

Moderation of soil disturbance following tillage operations using external inputs, particularly organic manures, is becoming an important aspect of soil tillage management. This study examined the effects of tillage, poultry manure (PM) and NPK fertiliser on soil physical properties and their relationship with maize yield. The experiment was a split-split-plot design with three replications. Tillage (zero tillage (ZT), minimum tillage (MT) and conventional tillage (CT)) was the main plot, the subplot contained four NPK fertiliser treatments (0, 90, 120 and 150 kg N ha–1), and the sub-subplots comprised three PM treatments (0, 10 and 20 Mg ha–1). The soil texture was gravelly loamy sand. Soil bulk density (BD), total porosity (Pt), clay dispersion ratio (CDR) and saturated hydraulic conductivity (Ks) were evaluated. Samples for BD and Ks were collected from the soil surface using stainless cylindrical soil core (5 cm in diameter and 5 cm in height), whereas samples for CDR determination were collected at 0–20 cm soil depth. Compared with the ZT, significantly lower BD was obtained under CT plots at 4 weeks after planting (WAP), which increased at 12 WAP possibly due to slaking and collapse of soil particles by raindrops. Application of PM resulted in significantly lower BD than the unmanured plots. There was a significant interaction effect among tillage, PM and NPK fertiliser on soil physical properties. In plots treated with combined 20 Mg ha–1 PM (PM20) and 150 kg N ha–1, the soil Pt of 55.6% and 57.2% obtained under CT and MT respectively at the end of the second year were significantly higher than those of the control plots (48.5% (CT) and 47.5% (MT)). In general, combined application of NPK and PM did not result in significant differences in CDR and Ks under MT and CT but the higher rate of PM (PM20) and NPK fertiliser under ZT significantly raised the CDR and reduced Ks by 20–43% in 2008 and 23–45% in 2009 relative to the unmanured plots. This implies disaggregation (higher CDR), which possibly resulted in creation of water-repellent properties causing lower Ks. Although Ks and CDR showed no significant relation with grain yield, BD was significantly related with maize grain yield with MT tending to produce ~23% and 20% higher maize grain yield than ZT and CT respectively under a typical arable surface soil BD. Therefore, integrated use of PM and NPK fertiliser is beneficial to soil physical qualities, especially under tillage management, but high quantities of PM should be avoided under ZT soil.

This study assessed the perception and use of digital applications for soil fertility management and conservation strategies among small-scale crop farmers in southwest Nigeria. A total of 376 farmers were randomly selected across the six southwest states. The data collected were analyzed using descriptive statistics. The majority of the farmers relied on perception and other non-scientific approaches such as the appearance of weeds and performance of crops in the previous season to assess soil fertility. Only 1.1% and 0.3% of the farmers assessed soil fertility through soil tests and digital applications, respectively. Most farmers adopted bush fallowing and the use of inorganic fertilizers to improve soil fertility. Although 4.8% of the farmers indicated that they had digital applications on their mobile phones, only 2.9% claimed to have used these. More than half (56.4%) of the farmers stated that a lack of awareness of the existence of digital applications and internet-enabled telephones were the reasons they have not been able to use digital applications. The majority of the farmers (97.3%) indicated their willingness to embrace the use of new farm decision digital applications which could provide more information, especially on soil fertility, if introduced. More extensive services focusing on older, less literate farmers and farmers who hitherto did not belong to any farmers’ association are advocated for in order to encourage the use of digital applications and soil fertility management and conservation practices.

The cultivation of cassava has been attributed to soil degradation through soil loss by water erosion. However, the magnitude of soil loss due to harvesting cassava (Manihot esculentus) is understudied. This study assessed the amount of soil and nutrient losses due to cassava manual harvesting from farmlands in Southwest Nigeria and investigated the factors that affect soil losses. Cassava harvesting was conducted twice at 8 locations in southwest Nigeria. The plant and soil samples were analyzed for planting density, root yield, soil texture, soil moisture, organic carbon, nitrogen, available phosphorus and potassium contents. The average soil loss from cassava harvesting was estimated as 8.35 t ha⁻¹ harvest⁻¹. The mean nutrient loss from the field was 9.95 kg N ha⁻¹ harvest⁻¹, 0.10 kg P ha⁻¹ harvest⁻¹ and 0.61 kg K ha⁻¹ harvest⁻¹. Clay content, soil moisture content, and root yield correlated positively with crop specific soil loss with correlation coefficients of 0.12, 0.53, and 0.57, respectively. Soil moisture and root yield are important factors that contribute to soil loss due to cassava harvesting. The high amount of soil and nutrient losses suggest that cassava harvesting could contribute to soil degradation in areas with high cassava root production. Therefore, further investigation should be conducted on harvesting procedures to minimize soil losses.

The current climate change scenario calls for efficient management of soil available water which requires adequate knowledge of soil hydraulic properties. This study was, therefore, carried out to evaluate the effect of tillage and mulching on soil hydraulic properties in a rice- wheat system. The study was a splitsplit plot experiment in randomized complete block design with three replications. The main plot was tillage under wheat cropping (conventional tillage (CTW) and zero tillage (ZTW)), the sub-plot involved tillage under rice cropping (CTR, ZTR and puddling (PTR)) while the sub-sub plot included rice straw mulching (0 and 6 t ha^-1). Soil volumetric water content (θ_v) was determined using time domain reflectometer. Tensiometers were installed to guide in field irrigation.Water evaporated from the soil and soil moisture enrichment (SME) were estimated using data obtained from the installed minilysimeters. It was observed that ZTW plots had higher θ_v than CTW plots similar to higher θ_v in mulched than unmulched plots. Tensiometric regulation of irrigation led to about 60% water saving without reduction in grain yield. Poor wheat tillering possibly increased soil water evaporation in ZTW than CTW plots while more water availability under CTR could be linked with more evaporation recorded in CTR than ZTR plots. Covering the soil surface with mulch led to about 154% reduction in soil evaporation compared with unmulched plots. Soil moisture enrichment by ZT was about 35% to 312% higher compared with CT while mulching raised SME by 205% compared with the unmulched soil. Insignificantly lower grain yield was obtained in ZTW plots than CTW while significantly (P<0.05) lower grain yield was recorded in CTR plots than ZTR. Therefore, appropriate tillage system coupled with soil surface management and monitoring of soil hydraulic properties during wheat growth are important to improve irrigation water efficiency and land productivity.