To solve fertility problems, most smallholder farmers in sub-Saharan Africa use fallow periods. However, population growth along with land shortage tend to shorten the duration of fallows, resulting in a steady decline in soil fertility. Assuming that nitrogen (N) plays a key role in soil fertility, we designed an ecological model describing N cycle in a cropping system. We examined the impact of different processes involved in N cycle, including mineralization, nitrification and fallow characteristics on the yield of a maize crop in a humid savanna, Côte d’Ivoire. The objective of this study was to explore ways to maintain N supply in N poor soils by identifying the appropriate levers and practices. The model revealed that in low input agricultural systems, soil fertility is maintained by the dynamics of soil organic matter and mineralization. We showed that, variation in nitrification during the cropping cycle (fallow-crop) does not have a significant effect on maize yield. However, with the addition of N fertilizers, reduced nitrification significantly increases crop yield. Indeed, low nitrification increases the efficiency of fertilizer use, which reduces the negative impact of excessive N fertilizer application. Furthermore, legume-based fallow was able to increase maize productivity much more than a nitrification-inhibiting fallow regardless of long duration of fallow periods. Also, the models suggested suggest that using nitrification-inhibiting grasses as cover crops for maize would be beneficial if mineral N fertilizer is used.
Research Article
The potential role of biological inhibition of nitrification in a fallow cropping system: a modelling approach
https://doi.org/10.21203/rs.3.rs-4177771/v1
This work is licensed under a CC BY 4.0 License
You are reading this latest preprint version
soil fertility
nitrification inhibition
nitrate
fallow
maize
small-holder agriculture
Population projections indicate that Africa will experience the highest population growth rate of any continent (Abramova 2022), which would require a strong increase in food production. However, we are witnessing a drop in crop yields due to decreasing soil fertility (Vanlauwe et al. 2015). In sub-Saharan Africa, soils have generally a low fertility and agricultural practices may lead to a decline in their productive capacity (Sheahan and Barrett 2014; Vanlauwe et al. 2015). To address soil fertility problems, most traditional subsistence cropping systems include a cropping phase followed by a fallow period, after observing a decline in yield. During fallows, grasses and forbs first grow, but shrubs or trees may grow in case of long-term fallows (more than 10 years). During the fallow, plants improve soil fertility through inputs of organic matter (both roots and leaves) and because no biomass is exported. This leads to an increase in the soil organic matter concentration, which increases the retention of mineral nutrients and water. This organic matter is also a source of fertility because it contains mineral nutrients that can be progressively released through mineralization (Johnston et al. 2009).
Depending on its duration and the type of plant community involved, fallows can be efficient at restoring soil fertility in low input cropping systems. Despite, the effectiveness of this type of cropping system (Nielsen et al. 2011), the rapid human population growth leads to a decrease in the per capita area of fields (Gafsi 2006) resulting in a decrease in fallow duration and consequently to a decline in soil fertility in sub-Saharan Africa (Faure 2005; Kopittke et al. 2019). In order to increase crop yields, farmers may turn to mineral fertilizers, which is potentially an efficient solution for West Africa (Falconnier et al. 2023) but requires financial means that are not affordable by resource-poor farmers. Moreover, the use of mineral fertilizers could have adverse environmental consequences that can strongly undermine the sustainability of agriculture (Kopittke et al. 2019). To maintain the fertility of these cropping systems and make them sustainable, it seems interesting to explore practices that are accessible to smallholder farmers. The objective of the current modelling study was thus to analyze practices and underlying factors that likely influence the long-term N fertility of these traditional fallow-based cropping systems and culture yield. For cultivation, we focused on maize because it is one of the most widely grown cereals in West Africa, especially in Côte d’Ivoire.
The accumulation of soil organic matter and the subsequent release of mineral N from this source of nutrients during cultivation is probably critical for the efficiency of the fallow (Johnston et al. 2009). We thus studied the impact of organic matter accumulation and the subsequent N mineralization on the yield and its dynamics along cropping cycles. Mineralization provides mineral N, directly available for plants, which immediately increases crop growth. However, this decreases future possibilities of mineral N release, which likely decreases future yields. We thus tested the impact of different mineralization rates on the maize yield. We also tested the impact of different fallow length and the number of cropping cycles between fallow periods on the subsequent yield.
Two forms of mineral N, nitrate (NO3−) and ammonium (NH4+), coexist and both are absorbed by plants (McFee and Stone 1968). Nitrification is the microbial process by which NO3− is produced from NH4+. Contrary to NH4+, NO3− is much more mobile, it is indeed easily leached and can be lost by denitrification, which respectively leads to the pollution of aquatic ecosystems by NO3− and emissions of N2O, a powerful greenhouse gas Thus, nitrification leads to losses of N fertility and, overall, decreases agriculture sustainability. One way of dealing with this issue is to decrease nitrification through ecological processes. This has been showed several years ago by Lata et al. (2004) and Subbarao et al. (2012) that some tropical savanna grasses inhibit nitrification through the release of root exudates. This decreases the availability of NO3−, which subsequently reduces losses of NO3−,and finally improves N conservation in ecosystems and thereby increases primary production (Boudsocq et al. 2009). The exploitation of nitrification inhibition, either by selecting nitrification-inhibiting crop varieties or through the use of nitrification-inhibiting grasses as cover crops, is consequently viewed as an important opportunity to increase agriculture sustainability both through the increase in N fertilizer efficiency, the reduction in pollution by NO3− and N2O emissions (Subbarao et al. 2006; Coskun et al. 2017). To test this hypothesis in the case of traditional fallow-based cropping systems, we modelled the effect of decreasing nitrification rates during the fallow and during the cropping cycles. Because the use of a legume-based fallow increasing fertility through N fixation has already been shown to be an efficient practice (Tondoh et al. 2013; Maseko et al. 2020), we compared it to the nitrification-inhibiting fallow.
Nitrification rates interact with plant preference for NO3− or NH4+ which could have a significant effect on crop productivity (Boudsocq et al. 2012; Konaré et al. 2019). For example, the conjunction of a high nitrification rate with a high preference for NH4+ likely leads to a high availability of NO3−, which would result in important N losses and a quick decrease in soil fertility. However, low nitrification rate with high preference for NH4+ would increase the efficiency of NH4+ recycling, thereby increasing productivity and reducing N losses (Boudsocq et al. 2009).
Based on the use of a simple and generic mathematical model representing N dynamics, our study aimed at evaluating the respective influence of the main possible levers improving soil fertility in fallow-based cropping systems. The model was parameterized for the region of Lamto humid savanna in Côte d’Ivoire, where the nitrification inhibition process of Hyparrhenia diplandra has been discovered. Overall, we tested (i) the effect of organic matter accumulation and subsequent mineralization on maize yield, (ii) the effect of nitrification during the cropping cycle in interaction with maize preference for NO3− versus (vs.) NH4+ and the potential use of mineral fertilization, (iii) the interaction between the durations of fallows and cropping cycles, (iv) the possibility to use an improved fallow either based on legumes or nitrification-inhibiting grasses.
2.1 Model description
We designed a mean-field model describing N dynamics in a traditional maize cropping system derived from the work of Konaré et al. (2019). The model includes four compartments: the biomass compartment (B), which represents either plant (a mixture of herbaceous, semi ligneous and ligneous plants depending of the stage of the fallow) biomass during the duration of the fallow or maize biomass during cultivation, the soil organic matter (O) compartment, and the two inorganic compartments, NH4+ (NA) and NO3- (NN) that represent the two main N sources (Fig. 1). The model compartments are N stocks expressed in kilograms of N per hectare (kg N ha-1) and the exchange rates between compartments are N fluxes, expressed in kilograms of N per hectare per day (kg N ha-1 d-1). Thus, the plant biomass refers to the size of the N stock in the plant compartment. All parameters describing N fluxes among the compartments are considered as constant values (Table 1).
Plants grow by taking up N from the inorganic compartments (NA and NN), according to their preference (β) for NO3- vs. NH4+ and with an uptake rate µ. Plant preferences for NH4+ and NO3- is represented by β1 and β2(Fig. 1), respectively, with β2equal to 1-β1. The preference for NH4+ (β1) is between 0 and 1. The closer the β1preference is to 1, the higher the absorption of NH4+. The same is true for the preference for NO3- (β2). This leads to the following expressions µBβ1NAfor NH4+ uptake and µBβ2NNfor NO3- uptake. The mortality of Biomass (B) with a rate d leads to a flux of dead organic matter towards the organic matter compartment (O). Soil organic matter is mineralized at a rate m leading to a N flux to the NA compartment. NH4+ can be transformed into NO3- by nitrification at a rate n. Overall, N uptake fluxes are controlled linearly by the donor and recipient (i.e., proportional to the size of the donor and recipient) but all other fluxes are modeled as linearly controlled by the donor.
We also considered in the model the fluxes of nutrients in and out of our compartments. The losses from the B, O, NA and NN compartments are represented by the parameters lB, lO, lNA, lNN respectively. In savanna, losses of N contained in biomasses are mainly due to fire, but during fallow periods and cropping cycles, there is no fire, which reduces losses. Losses of organic matter from soils are caused by erosion and leaching. Losses from the NH4+ compartment (lNA) are mainly due to volatilization, while losses from the NO3- compartment (lNN) are due to leaching and denitrification. The model also accounts for inputs of N to the agroecosystems. Wet and dry deposition provide constant inputs of N to the organic (RO) and inorganic (RNA, RNN) compartments. Taken together, the dynamics of N in the different compartments are expressed by the following system of differential equations:
2.2 Parameterization
The values of the different parameters used are summarized in Table 1. The generic N parameter values were derived from previous work done in Lamto savanna (Villecourt and Roose 1978; Boudsocq et al. 2009; Konaré et al. 2019). However, yearly parameters were divided by 365 as the model is parameterized for a daily basis (Table 1). To cope with the seasonality of Lamto climate known to be characterized by four seasons including two rainy seasons and two dry seasons, two maize cropping cycles were modeled per year to match agricultural practices around Lamto. The first cycle takes place during the first 105 days (main rainy season from May to mid-August) while the second cycle begins during the second rainy season and has the same duration (105 days, from mid-August to November). The remaining 5 months (155 days) correspond to the dry season where normally no cropping activities are carried out. During these months, weeds grow and just before the beginning of the next rainy season, the soil is weeded, which leads to an input of dead organic matter to the soil for the next crop cycle. After each maize harvest, the biomass of grain is removed from the total biomass as output while the remaining biomass is returned to the soil as organic matter. Studies conducted by Yakoub (2015), showed that approximately 30% of the total N is in the maize grain. Thus, at each harvest, 30% of the grain biomass, considered as the crop, is removed from the total biomass and exported. The remaining 70% is returned to the soil. As an annual plant, the turnover rate of maize biomass (d) during its growth is very low and losses of its biomass (lB) are null. The values of the uptake rate and preference (β1 and β2) of maize were determined by varying these two parameters to obtain a realistic maize biomass,300 kg N ha-1, that is consistent with the literature (Yakoub 2015). During the fallow period, N cycle functions in the same way as during the cropping cycle. However, the recycling rate (d) during fallow is much higher than that of the crop cycle. The absence of fire during the fallow period reduces fire-related losses of dead organic matter or plant biomass (lO, lB). The biomass harvested at the end of the fallow is returned to the soil. To obtain a fairly realistic amount of biomass during the fallow period, fallow parameters (d, m, β, and u) were adjusted (table 1) to reach a biomass of a 10 years fallow (Hauser and Nolte 2002).
2.3 Numerical simulations
The model was numerically solved through the R software (R Development Core Team 2022) and the simulations were performed for daily time steps. The numerical resolution of the differential equations was performed using the deSolve package (Soetaert et al. 2010).
The first simulations model cropping cycles without periods of fallows and consisted in testing the effects on N cycling and maize yield of the interaction between (i) the mineralization rate (three different mineralization levels: low mineralization = 1.10-4 day-1, medium mineralization = 5.10-4 day-1 and high mineralization = 9.10-4 day-1), (ii) the rate of absorption of mineral N ranging between 0 and 0.01 ha kg N-1 day-1; and (ii) the preference for NH4+ vs. NO3- ranging from 0 to 1. Simulations were started with a dead organic matter stock of 2,5.103 kg N ha-1. Additionally, we conducted simulations to investigate how the duration of fallow (5, 10 and 20 years) and cultivation (1, 3 and 5 years) interacted to influence maize yield. We compared the maize productivity after three fallow types – N-fixing fallow, simple fallow, and nitrification-inhibiting fallow. We finally applied increasing doses of fertilizer (0, 10, 50, 100, 200 kg N ha-1) at two nitrification rates, low vs. high, to determine the effect of nitrification inhibition during the cropping cycles on maize productivity.
3.1 Results
During a one-year cropping cycle, maize grew by uptaking the mineral N that is mineralized from the soil organic matter. The maize biomass increases with the absorption rate and is maximum for intermediate preferences for NO3− vs. NH4+ that are higher than 0.5 (preference for NH4+) (Fig. 2). The increase in maize biomass with the uptake rate increases steeply for low uptake rate values and then reaches a plateau. This general pattern is the same for the three mineralization rates (Fig. 2), but the maximum maize biomass increases from less than 100 kg N ha− 1 for the lower mineralization rate, to less than 300 kg N ha− 1 for the intermediate mineralization rate, to less than 500 kg N ha− 1 for the higher mineralization rates. The positive effect of the uptake rate on maize biomass is steeper for high mineralization rates.
Fallow periods allow for the constitution of important reserve of soil organic matter (and the N it contains) (Fig. 3). This reserve begins to decrease as soon as maize is sown due to mineralization, and maize grows absorbing the resulting mineral nitrogen. The soil is unable to fully regain its fertility between the two cropping cycles of the same year so that the yield of the second cropping cycle is lower than the first one. The biomass obtained from the first crop depressed from 185 kg N ha− 1 to 178 kg N ha− 1 for the second crop (Fig. 3). for the second crop. This is due to the quick mineralization of the soil organic matter stored during the fallow and the fact that the maize biomass (leaves and roots) added to the soil after the first maize cycle is not large enough to reach the quantity of soil organic matter reached at the end of the first fallow year.
Overall, maize biomass overall increased with the duration of fallows and decreased during successive years of maize cultivation (Fig. 4). There is an interaction between fallow duration and the number of successive years of cultivation. Hence, for one year of cultivation the yield remains high for successive fallow cycles and even increases for longer fallows. For 3 years of successive cultivation, the maize biomass decreases between fallow cycles for the shorter fallow, i.e. fallows no longer maintain fertility on the long term. For five years of successive cultivation, the yield decreases between fallow cycles whatever the length of the fallow.
Corn biomass varies greatly with the addition of N fertilizer and the effect of nitrification (Fig. 5). Without N fertilizer, the biomass of the two crops is almost identical. When we add fertilization (200 kg N ha− 1), we obtain for a crop under high nitrification a biomass of 242 kg N ha− 1 while the biomass under low nitrification is approximately 385 kg N ha− 1. We note that, maize biomass increases with the rate of N fertilizer application (Fig. 5). However, this increase is much stronger for a low nitrification rate during the cultivation than for a high nitrification rate. In other words, without mineral fertilization, low nitrification does not impact maize biomass, but the positive effect of low nitrification increases with the fertilization rate.
The legume-based fallow (improved fallow) has a much higher maize biomass than the other two fallow types, regardless of fallow duration (344, 450 and 657 kg N/ha respectively for fallow durations of 5, 10 and 20 years). The two other fallow types, resulted in almost similar maize biomass: low-nitrification fallow did not increase maize biomass sufficiently compared with the standard fallow (Fig. 6). For a 5-year fallow, we obtain a yield of 172 kg N/ha for low-nitrification and simple fallows. This yield increases slightly for a 10-year fallow (185 kg N/ha) and reaches 211 kg N/ha for a 20-year fallow
3.2 Discussion
Our results emphasized the particular influence of some of the mechanisms involved in N cycling, which overall confirms our hypotheses: (i) organic matter dynamics and mineralisation rates have a determining impact on soil fertility whatever the preference for NH4+ vs. NO3−; (ii) the inhibition of nitrification during the crop cycle increases the yield, but only in case of mineral N fertilization; (iii) long term fallow allows the building up a N reserve within soil organic matter, which maintains fertility as long as the fallow is long enough relatively to the number of successive cropping cycles; (iv) legume-based fallow increases soil fertility and crop yield much more than a nitrification-inhibiting one.
The initial results of the model confirmed the critical role of organic matter contributes in soil fertility replenishment and maintenance. According to Biaou et al. (2018), incorporating organic matter into the soil enhances its fertility through various mechanisms, one of which being considered in our model, i.e. the increase in the organic stock of mineral nutrients. Furthermore, organic matter mineralization rate plays a crucial role in enhancing maize yield (Thiebeau and Recous 2017). Current fertility requires a large enough stock of organic N and a sufficient instantaneous mineralization of this stock. Thus, in smallholder cropping systems with no mineral fertilization, soil fertility indeed depends on the past dynamics of organic matter accumulation during which inputs of organic N are higher than losses of N due to mineralization. This is partially due to mineral N being much more prone to losses than organic N.
Figure 2 clearly revealed that an increase in mineralization rate leads to a corresponding increase in maize yield, but only if a large soil organic matter stock has built up beforehand. The combined influence of absorption rate and NH4+ preference on maize growth biomass decreases with an increase in mineralization. In cases where the absorption rate is very high, the quantity of organic matter and mineralization rate become the decisive factors for successful cultivation of maize. Our results are in accordance with the empirical study of Cobo et al. (2002) that showed a strong correlation between organic matter decomposition rate, nutrient release, and the subsequent increase in yield due to nutrient absorption by crops. However, it is worth noting that mineralization, nutrient absorption and crop exportation necessarily gradually decrease fertility in a system without fertilization.
The preference for NH4+ vs. NO3− did not strongly impact maize biomass (Fig. 2) within a large range of preferences around 0.75 (i.e. 75% of N absorption as NH4+). Nevertheless, the yield strongly decreases for extreme values of preference and the lower the mineralization and the higher N absorption the narrower the range of preferences allowing for the maximal yield. These results are supported by previous modelling results (Boudsocq et al. 2009) showing that intermediate preferences for NO3− vs NH4+ with a preference for NH4+ increases primary production. Indeed, absorbing both NO3− and NH4+ prevents the crop to deprive itself of one of the two resources, and absorbing a little more NH4+ tends to reduce nitrification flows and subsequent losses by leaching and denitrification (Boudsocq et al. 2012; Konaré et al. 2019). While our study was based on constant preference values, it is important to acknowledge that the plant preference for a particular form of mineral N depends on many factors. Defining plant preference for NH4+ vs. NO3− can indeed be challenging, as it hinges on interactions between multitude of complex and dynamic environmental and physiological factors (Britto and Kronzucker 2013). These include soil chemistry, temperature, pH levels, plant genotype and stage of plant development (McFee and Stone 1968). Wang et al. (2021) demonstrated that maize prefers NH4+ in acidic soils, whereas Zhang et al. (2019) found that maize prefers NH4+ in neutral soils (subtropical humid monsoon climate). Nevertheless, further research is required in this area to better understand plant preference between NO3− and NH4+.
Like in past studies, our results confirmed that long fallow periods are required to maintain fertility in the absence of fertilization. For example, for 5 successive cropping cycles involving at least 20 years fallows are necessary for the N sustainability of the production. This is indeed due to the fact that exportations of N with the crop must be compensated, and that fallows allow for building up again fertility by stopping exportation and accumulating atmospheric inputs of N through dry and wet depositions that are absorbed by fallow vegetation and accumulate in the soil as organic N. Overall, our results confirm finding from empirical studies suggesting that the duration of fallow periods is important in restoring soil fertility (Atchada et al. 2019). In shifting cultivation systems, fallow duration significantly influences soil and vegetation dynamics, as observed in Fig. 4, which revealed that maize yield increases with the length of fallow periods (Toky and Ramakrishnan 1981; Swamy and Ramakrishnan 1988). However, the effectiveness of fallows decreases as cropping periods extend. To maintain the positive impact of fallows, it is necessary to increase the duration of the fallow period. This is confirmed by Kouelo (2015) who recommended a minimum fallow period of 30 years to restore degraded soils in sub-Saharan Africa. However this seems unrealistic as the need to satisfy food security of the growing population (Gafsi 2006; Abramova 2022) in sub-Saharan Africa makes it difficult to maintain such a long fallow period.
In our model, low nitrification during the cropping cycles increases maize yield only when mineral fertilization is used and this effect increases with the amount of N fertilizer (Fig. 5). In the absence of mineral fertilization, the possibility of NO3− leaching is low as maize absorbs a significant amount of NH4+ and NO3− so that the amount of available mineral N remains low all time. In this case, the nitrification rate (low or high nitrification) has minimal impact on maize biomass (Fig. 5). However, with inputs of mineral N fertilizer, the stock of mineral N increases transiently. In this case, a low nitrification rate reduces the quantity of NO3−, thereby minimizing losses through leaching and denitrification. These reduced N losses increase the availability of mineral N, in the form of NH4+, which promotes better plant growth and increases maize biomass. These results confirmed empirical results (Hu and Schraml 2014; Zhang et al. 2015a; Karwat et al. 2017) that showed that synthetic nitrification inhibitors with fertilizers reduce N losses and increase N availability, while maintaining good yield (Hu and Schraml 2014; Muller et al. 2023). However, our results seem to contradict findings by Boudsocq et al. (2012) and Konaré et al. (2019) suggesting that the inhibition of nitrification by savanna grasses strongly increases grass biomass production without any human fertilization. Two mechanisms likely explain the apparent discrepancy between these modelling results. First, our study focuses by definition on the transient dynamics of N cycling during a few cultivation cycles, whereas Boudsocq et al. (2012); Konaré et al. (2019) focused on long-term dynamics of ecosystems (equilibrium property of models). Such transient dynamics probably do not let enough time for the building up of N stocks through the repeated reduction of losses of NO3−. Second, the relatively high productivity of maize modelled requires a high rate of mineral N uptake by the crop. This reduced the availability of mineral nitrogen without fertilization, which may contribute to the lack of effect of low nitrification (Fig. 5).
According to Subbarao et al. (2012), sustainable agriculture requires the development of production systems incorporating new crop varieties capable of regulating nitrification and improved agronomic practices to reduce the leakage of N throughout the N cycle, a critical requirement to increase food production sustainably while mitigating adverse environmental effects. While inhibiting nitrification is viewed as very promising avenue to increase agriculture sustainability (Subbarao et al. 2012; Coskun et al. 2017), our model revealed some limits to these prospects. Rather logically, the availability of mineral NH4+ should be high enough and during a period sufficiently long for the nitrification inhibition to have an impact on yield. We also suggest that reduced nitrification would be more effective in agricultural systems with high mineral N inputs than in traditional subsistence agriculture without mineral fertilization.
Our results showed that the best way to improve the effect of fallows on N fertility is to grow N-fixing legumes in the fallow (Fig. 6). This confirms already well-established empirical results by Soro et al. (2015) that demonstrate the advantages of legume-based fallows even with short fallows (Williams et al. 2022). On the contrary, implementing a low-nitrification fallow does not significantly impact N fertility. This is likely explained, as for the low effect of a low nitrification rate during cropping cycles (see above), by the low availability of mineral N during the fallow (results not show). The impact of nitrification inhibition during fallow on yield and fertility remains largely unexplored experimentally. However, our results seem to be in contradiction with results showing that perennial savanna grasses, improve fertility probably through the inhibition of nitrification (Yé et al. 2017). A possible explanation is that we have used in our simulations a high value for mineral N absorption rate by fallow vegetation, while a lower absorption would increase mineral availability and increase the impact of reducing nitrification. Investing further this issue would require better describing N dynamics and the availability of NO3− and NH4+ during fallows.
Most agricultural system models are complex models that provide synthesis and quantification to assess the effects of water, soil, crops, trees, management practices and climate on the sustainability of agricultural production and to ensure food security (Dupraz et al. 2019; Burgess et al. 2019; Ahmed et al. 2022). On the contrary, our study was carried out using a fairly simple yet very general model, offering significant flexibility to consider key factors influencing N dynamics within the ecosystem. It is important to note that the model is not intended to lead to precise quantitative predictions, but rather to enhance our understanding of the system and to allow general predictions that are relevant to agriculture. To enhance the quantitative predictive accuracy of the model, we may consider incorporating additional compartments and mechanisms. For example, the absorption of mineral N, and the underlying complex plant-soil interactions, could be modelled using a function allowing for non-linearity between absorption, mineral stocks and plant biomass. Similarly, the preference for NO3− vs. NH4+ is probable not constant and influenced by many factors (Britto and Kronzucker 2013). These mechanisms could also be added in the model. Nevertheless, the generality of our model guarantees that our results are likely applicable to other crops and many fallow-based subsistence cropping systems from around the world. Up to our knowledge, it is also the first modelling effort analyzing the potential effects of biological nitrification inhibition on the sustainability of agriculture, in a context where N fertility and fertilization are at the heart of many sustainability issues (Zhang et al. 2015b).
By analyzing the influence of several ways of acting on N fertility, our study confirms that fallows are an effective system to maintain soil fertility. However, they must be long enough to maintain their benefits. Yet, fallow duration tends to decrease in Africa (Aihou 2003; Ouédraogo 2004; Gaiser et al. 2011) due to human population increase, food demand and land pressure increase. Some studies emphasized the necessity to increase the use of mineral fertilizers in Africa (Falconnier et al. 2023), which would be technically a good solution. However, one should not overlook the economic and environmental implications. Indeed, it is difficult for farmers in subsistence agriculture to afford buying fertilizers. This could become possible if fertilizers increase yields, which would allow farmers to sell a part of their production to the growing proportion of city inhabitants in Africa (Ciceri and Allanore 2019). Nevertheless, producing N fertilizer relies on the use of fossil fuels, which contributes to global warming, and the use of N fertilizers tends to increase NO3− leaching threatening aquatic ecosystem through eutrophication (Ansari et al. 2011) and leading to denitrification contributing to global warming through the production of N2O (Ravishankara et al. 2009). This has led to the crossing of the planetary boundary corresponding to N (Richardson et al. 2023) especially because of intensive agriculture in temperate areas,. Thus, global sustainability requires decreasing the use of N fertilizers and increasing this use in Africa will reinforce this problem
Our results suggested that decreasing nitrification during cropping cycles could increase the efficiency of N absorption and fertilizers, decrease the amounts of N fertilizers and decrease the subsequent environmental negative impacts of fertilization. This could be achieved using artificial nitrification inhibitors (Hu and Schraml 2014; Cui et al. 2022; Muller et al. 2023) through the use of nitrification inhibiting grasses as cover crops (Bozal-Leorri et al. 2023). This solution which seems promising in Africa where nitrification inhibiting grasses originate, remains to be developed. Selecting new crop, maize in our case, varieties that inhibit nitrification is also viewed as a promising possibility (Otaka et al. 2022; Petroli et al. 2023). Our results also confirm that using legume-based fallows would be an alternative solution to the increase in N fertilizer use. This practice is so far not generalized in Africa and not implemented in humid savanna areas for which our model has been parameterized.
Funding: The study was funded by the International Doctoral Program Modelling of Complex Systems (PDI MSC) funded by the Institute of Research for Development (IRD).
Author’s contributions: All authors contributed to the development of the model. The first draft of the manuscript was written by Waogninlin Amed Ouattara, with the feedback from Sarah Konaré, Ebagnerin Jérôme Tondoh and Sébastien Barot. All authors read and approved the final manuscript.
Code availability: Model simulation were conducted with R. The packages are mentioned in the “Method” section.
Availability of data: This a modelling paper. All data regarding the model are included in the paper.
Conflicts of interest: The authors declare no competing interests.
Ethic approval: Not applicable.
Consent to participate: Not applicable.
Consent for publication: Not applicable.
- Abramova IO (2022) The Population of Africa under the Conditions of Transformation of the World Order. Her Russ Acad Sci 92:S1306–S1315. https://doi.org/10.1134/S1019331622200023
- Ahmed Z, Gui D, Qi Z, et al (2022) Agricultural system modeling: current achievements, innovations, and future roadmap. Arab J Geosci 15:363. https://doi.org/10.1007/s12517-022-09654-7
- Aihou K (2003) Interaction between organic input by Cajanus cajan (L.) Millsp. and inorganic fertilization to maize in the derived savanna of the Benin Republic /. https://lib.ugent.be/catalog/rug01:000777292
- Ansari AA, Singh Gill S, Lanza GR, Rast W (eds) (2011) Eutrophication: causes, consequences and control. Springer Netherlands, Dordrecht
- Atchada C, Zoffoun A, Akplo T, et al (2019) Modes d’utilisation des terres et stock de carbone organique du sol dans le bassin supérieur de Magou au Bénin. Int J Biol Chem Sci 12:2818. https://doi.org/10.4314/ijbcs.v12i6.27
- Biaou ODB, Saidou A, Bachabi F-X, et al (2018) Effet de l’apport de différents types d’engrais organiques sur la fertilité du sol et la production de la carotte (Daucus carota L.) sur sol ferralitique au sud Bénin. Int J Biol Chem Sci 11:2315. https://doi.org/10.4314/ijbcs.v11i5.29
- Boudsocq S, Lata JC, Mathieu J, et al (2009) Modelling approach to analyse the effects of nitrification inhibition on primary production. Funct Ecol 23:220–230. https://doi.org/10.1111/j.1365-2435.2008.01476.x
- Boudsocq S, Niboyet A, Lata JC, et al (2012) Plant Preference for Ammonium versus Nitrate: A Neglected Determinant of Ecosystem Functioning? Am Nat 180:60–69. https://doi.org/10.1086/665997
- Bozal-Leorri A, Corrochano-Monsalve M, Arregui LM, et al (2023) Evaluation of a crop rotation with biological inhibition potential to avoid N2O emissions in comparison with synthetic nitrification inhibition. J Environ Sci 127:222–233. https://doi.org/10.1016/j.jes.2022.04.035
- Britto DT, Kronzucker HJ (2013) Ecological significance and complexity of N-source preference in plants. Ann Bot 112:957–963. https://doi.org/10.1093/aob/mct157
- Burgess P, Graves A, García de Jalón S, et al (2019) Modelling agroforestry systems. pp 209–238
- Ciceri D, Allanore A (2019) Local fertilizers to achieve food self-sufficiency in Africa. Sci Total Environ 648:669–680. https://doi.org/10.1016/j.scitotenv.2018.08.154
- Cobo JG, Barrios E, Kass DCL, Thomas RJ (2002) Decomposition and nutrient release by green manures in a tropical hillside agroecosystem. Plant Soil 240:331–342. https://doi.org/10.1023/A:1015720324392
- Coskun D, Britto DT, Shi W, Kronzucker HJ (2017) Nitrogen transformations in modern agriculture and the role of biological nitrification inhibition. Nat Plants 3:17074. https://doi.org/10.1038/nplants.2017.74
- Cui L, Li D, Wu Z, et al (2022) Effects of combined nitrification inhibitors on soil nitrification, maize yield and nitrogen use efficiency in three agricultural soils. PLOS ONE 17:e0272935. https://doi.org/10.1371/journal.pone.0272935
- Dupraz C, Wolz K, Lecomte I, et al (2019) Hi-sAFe: A 3D Agroforestry Model for Integrating Dynamic Tree–Crop Interactions. Sustainability 11:2293. https://doi.org/10.3390/su11082293
- Falconnier GN, Cardinael R, Corbeels M, et al (2023) The input reduction principle of agroecology is wrong when it comes to mineral fertilizer use in sub-Saharan Africa. Outlook Agric 52:311–326. https://doi.org/10.1177/00307270231199795
- Faure G (2005) Valorisation agricole des milieux de savanes en Afrique de l’Ouest : des résultats contrastés. Cah O-m 58:5–24. https://doi.org/10.4000/com.262
- Gafsi M (2006) Exploitation agricole et agriculture durable. Cah Agric 15:491–497. https://doi.org/10.1684/agr.2006.0035
- Gaiser T, Judex M, Igué AM, et al (2011) Future productivity of fallow systems in Sub-Saharan Africa: Is the effect of demographic pressure and fallow reduction more significant than climate change? Agric For Meteorol 151:1120–1130. https://doi.org/10.1016/j.agrformet.2011.03.015
- Hauser S, Nolte C (2002) Biomass production and N fixation of five Mucuna pruriens varieties and their effect on maize yields in the forest zone of Cameroon. J Plant Nutr Soil Sci 165:101.; https://doi.org/10.1002/1522-2624(200202)165:1<101::AID-JPLN101>3.0.CO;2-F
- Hu Y, Schraml M (2014) Influence of nitrification inhibitors on yields of arable crops: A meta-analysis of recent studies in Germany. Int J Plant Prod 8:33–50
- Johnston AE, Poulton PR, Coleman K (2009) Chapter 1 Soil Organic Matter: Its Importance in Sustainable Agriculture and Carbon Dioxide Fluxes. In: Sparks DL (ed) Advances in Agronomy. Academic Press, pp 1–57
- Karwat H, Moreta D, Arango J, et al (2017) Residual effect of BNI by Brachiaria humidicola pasture on nitrogen recovery and grain yield of subsequent maize. Plant Soil 420:389–406. https://doi.org/10.1007/s11104-017-3381-z
- Konaré S, Boudsocq S, Gignoux J, et al (2019) Effects of Mineral Nitrogen Partitioning on Tree–Grass Coexistence in West African Savannas. Ecosystems 22:1676–1690. https://doi.org/10.1007/s10021-019-00365-x
- Kopittke PM, Menzies NW, Wang P, et al (2019) Soil and the intensification of agriculture for global food security. Environ Int 132:105078. https://doi.org/10.1016/j.envint.2019.105078
- Kouelo FA (2015) Effects of Cropping Practices on Soil Degradation in Three Watersheds in Southern Benin [Effets des pratiques culturales sur la dégradation du sol au niveau des trois bassins versants du sud Bénin. Thèse de Doctorat en Sciences Agronomiques, Universités d’Abomey-Calavi
- Lata J-C, Degrange V, Raynaud X, et al (2004) Grass populations control nitrification in savanna soils. Funct Ecol 18:605–611. https://doi.org/10.1111/j.0269-8463.2004.00880.x
- Maseko ST, Maredi MP, Mathews C, Dakora FD (2020) Harnessing ecosystem services from biological nitrogen fixation. In: The Role of Ecosystem Services in Sustainable Food Systems. Elsevier, pp 73–94
- McFee WW, Stone EL (1968) Ammonium and Nitrate as Nitrogen Sources for Pinus radiata and Picea glauca. Soil Sci Soc Am J 32:879–884. https://doi.org/10.2136/sssaj1968.03615995003200060045x
- Muller J, De Rosa D, Friedl J, et al (2023) Combining nitrification inhibitors with a reduced N rate maintains yield and reduces N2O emissions in sweet corn. Nutr Cycl Agroecosystems 125:107–121. https://doi.org/10.1007/s10705-021-10185-y
- Nielsen UN, Ayres E, Wall DH, Bardgett RD (2011) Soil biodiversity and carbon cycling: a review and synthesis of studies examining diversity–function relationships. Eur J Soil Sci 62:105–116. https://doi.org/10.1111/j.1365-2389.2010.01314.x
- Otaka J, Subbarao GV, Ono H, Yoshihashi T (2022) Biological nitrification inhibition in maize—isolation and identification of hydrophobic inhibitors from root exudates. Biol Fertil Soils 58:251–264. https://doi.org/10.1007/s00374-021-01577-x
- Ouédraogo E (2004) Soil quality improvement for crop production in semi-arid West Africa. Wageningen University and Research Centre
- Petroli CD, Subbarao GV, Burgueño JA, et al (2023) Genetic variation among elite inbred lines suggests potential to breed for BNI-capacity in maize. Sci Rep 13:13422. https://doi.org/10.1038/s41598-023-39720-3
- R Development Core Team (2022) R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing. https://www.R-project.org
- Ravishankara AR, Daniel JS, Portmann RW (2009) Nitrous Oxide (N 2 O): The Dominant Ozone-Depleting Substance Emitted in the 21st Century. Science 326:123–125. https://doi.org/10.1126/science.1176985
- Richardson K, Steffen W, Lucht W, et al (2023) Earth beyond six of nine planetary boundaries. Sci Adv 9:eadh2458. https://doi.org/10.1126/sciadv.adh2458
- Sheahan M, Barrett CB (2014) Understanding the Agricultural Input Landscape in Sub-Saharan Africa : Recent Plot, Household, and Community-Level Evidence. The World Bank
- Soetaert K, Petzoldt T, Setzer RW (2010) Solving Differential Equations in R: Package deSolve. J Stat Softw 33:1–25. https://doi.org/10.18637/jss.v033.i09
- Soro K, Mangara A, Kouame NMT, et al (2015) Impact d’une jachère améliorée et du calendrier cultural sur le rendement du maïs dans le Sud de la Côte d’Ivoire. Int J Biol Chem Sci 9:1790. https://doi.org/10.4314/ijbcs.v9i4.5
- Subbarao G, Ito O, Sahrawat K, et al (2006) Scope and Strategies for Regulation of Nitrification in Agricultural Systems—Challenges and Opportunities. Crit Rev Plant Sci 25:303. https://doi.org/10.1080/07352680600794232
- Subbarao GV, Sahrawat KL, Nakahara K, et al (2012) Biological Nitrification Inhibition—A Novel Strategy to Regulate Nitrification in Agricultural Systems. In: Advances in Agronomy. Elsevier, pp 249–302
- Swamy PS, Ramakrishnan PS (1988) Ecological implications of traditional weeding and other imposed weeding regimes under slash-and-burn agriculture (jhum) in northeastern India. Weed Res 28:127–136. https://doi.org/10.1111/j.1365-3180.1988.tb00795.x
- Thiebeau P, Recous S (2017) Dynamiques de décomposition des résidus de cultures sur des exploitations pratiquant l’agriculture de conservation en région Grand Est, France. Cah Agric 26:65001. https://doi.org/10.1051/cagri/2017050
- Toky OP, Ramakrishnan PS (1981) Cropping and yields in agricultural systems of the north-eastern hill region of India. Agro-Ecosyst 7:11–25. https://doi.org/10.1016/0304-3746(81)90012-3
- Tondoh JE, Koné AW, N’Dri JK, et al (2013) Changes in soil quality after subsequent establishment of Chromolaena odorata fallows in humid savannahs, Ivory Coast. CATENA 101:99–107. https://doi.org/10.1016/j.catena.2012.10.005
- Vanlauwe B, Six J, Sanginga N, Adesina AA (2015) Soil fertility decline at the base of rural poverty in sub-Saharan Africa. Nat Plants 1:15101. https://doi.org/10.1038/nplants.2015.101
- Villecourt P, Roose E (1978) Charge en azote et en éléments minéraux majeurs des eaux. Rev Ecol Biol Sol 15:1–20
- Wang JL, Zhao XQ, Zhang HQ, Shen RF (2021) The preference of maize plants for nitrate improves fertilizer N recovery efficiency in an acid soil partially because of alleviated Al toxicity. J Soils Sediments 21:3019–3033. https://doi.org/10.1007/s11368-021-03007-9
- Williams A, Kay P, Stirling G, et al (2022) Impacts of reducing fallow periods on indicators of soil function in subtropical dryland farming systems. Agric Ecosyst Environ 324:107727. https://doi.org/10.1016/j.agee.2021.107727
- Yakoub AM (2015) Mineral Nitrogen Fertilization and Stover Management effects on Maize Production under Irrigated Mediterranean Conditions. Simulation of Yields. Universitat de Lleida
- Yé L, Lata J-C, Masse D, et al (2017) Contrasted effects of annual and perennial grasses on soil chemical and biological characteristics of a grazed Sudanian savanna. Appl Soil Ecol 113:155–165. https://doi.org/10.1016/j.apsoil.2017.02.003
- Zhang HQ, Zhao XQ, Chen YL, et al (2019) Case of a stronger capability of maize seedlings to use ammonium being responsible for the higher 15N recovery efficiency of ammonium compared with nitrate. Plant Soil 440:293–309. https://doi.org/10.1007/s11104-019-04087-w
- Zhang M, Fan CH, Li QL, et al (2015a) A 2-yr field assessment of the effects of chemical and biological nitrification inhibitors on nitrous oxide emissions and nitrogen use efficiency in an intensively managed vegetable cropping system. Agric Ecosyst Environ 201:43–50. https://doi.org/10.1016/j.agee.2014.12.003
- Zhang X, Davidson EA, Mauzerall DL, et al (2015b) Managing nitrogen for sustainable development. Nature 528:51–59. https://doi.org/10.1038/nature15743
Table 1 is available in the Supplementary Files section