Cocoa (Theobroma cacao Linn.) is an important cash crop in many tropical countries, given its role in their economies [1] . Africa still holds the monopoly on production, with over 70% of the world’s cocoa supply [2] . Increasing production has thus become a necessity and a challenge for the main producing countries [3] , including Côte d’Ivoire, the world’s leading producer since 1970, which produced 2.2 million tonnes of merchantable cocoa, according to the 2022-2023 campaign [4] . Revenues from cocoa sales generate 15% of GDP and account for over 50% of export earnings [5] . Cocoa cultivation covers more than 2 million hectares of land and employs 60% of the working population [6] .

However, cocoa production in Côte d’Ivoire, which is partly the result of attempts at intensification and replanting, but also of the creation of new plantations at the expense of forests, is facing a decline in yields due to a reduction in the country’s forest reserves [7] [8] , with the destruction of forest cover falling from 16 million hectares in 1960 to 3.4 million hectares in the last decade, the aging of orchards, diseases such as brown pod rot and swollen shoot [9] , and the rapid decline in soil fertility [10] due to overexploitation. In order to be self-sufficient in food, cocoa farmers cultivate various crops, including cocoa, yams, bananas, and sometimes taro and corn, on the same land for several years. This practice, which gives the soil no respite, depletes it of nutrients. To compensate for this and maintain a certain level of productivity, cocoa farmers invest in various inputs. This is also the conclusion reached by some authors when they assert that, after several unsuccessful attempts by cocoa farmers to boost their production, they have ended up relying on the use of fertilizers, particularly foliar fertilizers first and granular fertilizers second, in the hope of restoring the vegetative system of the cocoa trees [11] [12] . Fertilizer appears to be one of the factors that could potentially revitalize the economy of an old cocoa-growing region and curb migration to remaining forests. It can play this role locally and partially [13] .

But on a national scale, the logic of cocoa migration and the conquest of new forests has always prevailed. In fact, in the 2000s and 2010s, despite talk of “zero deforestation” and fertilizer as a tool for sustainability, migration and deforestation continued unabated [14] . Faced with this prevailing situation, the cocoa industry continues to advocate the use of mineral fertilizers as a sustainability tool to improve production. However, the criteria and techniques for assessing the fertility of cocoa-growing soils vary depending on their users. Some are generally based on the decrease in organic matter content, while others are based on the decline in marketable cocoa yields [15] . With a view to finding other solutions to improve cocoa production, this study, which aims to use mineral fertilizer combined with liquid fertilizer, was initiated to evaluate the effectiveness of these fertilizers.

Sustainable cocoa production over 25 to 30 years depends largely on soil fertility, hence the importance of mineral fertilization [19] . This fertilization intensifies nutrient levels, enabling high yields (over 2 tha⁻¹) to be obtained. It must be maintained to replace the nutrients exported by the crops [20] . Thus, the results of this study indicate that cocoa trees receiving certain combined doses of mineral and liquid fertilizers have overall good production characteristics. This is reflected, among other things, in treatments T4 (300 g NPK 0-23-19 + TE + 0.045 L NPK 11-9-41), T5 (300 g NPK 0-23-19 + TE + 0.09 L NPK 11-9-41) and T9 (150 g NPK 0-23-19 + TE + 0.18 L NPK 11-9-41), which were more effective in terms of wilted cherelles and actual and potential yields. As for other parameters, notably healthy, rotted and total pods and fruit weight, there were no significant differences between the fertilized plots and the controls, notably absolute (T0) and reference (NPK 0-23-19). The increase in yield obtained on the sites would therefore be linked to the nutrients contained in the solid and liquid fertilizers. The combination of solid and liquid fertilizers enabled the cocoa trees to better absorb the mineral elements needed to boost production. In fact, the phosphorus and potassium contained in solid and liquid fertilizers are recognized as major contributors to flower proliferation and improved fruit quantity and quality. Regular use of phosphorus-based mineral fertilizers is necessary to support canopy growth and bean production [21] . As for potassium, it plays a role in facilitating assimilate transport, improving stomatal function and photosynthesis [22] - [24] . Its role also contributes to improving production yields. Previous studies have shown the positive effect of potassium on the number of pods [25] . Studies have also shown that the number of pods is even more pronounced [26] when we have the combined effect of nitrogen, phosphorus and potassium fertilization. The positive effect of fertilizers on cocoa production can be explained by an improvement in the chemical status of the soil [27] . Research conducted at the station has also shown the importance of phosphorus and potassium in improving cocoa yields [28] . Some authors also claim that after several unsuccessful attempts by cocoa farmers to boost their production, they ended up relying on the use of fertilizers, particularly foliar fertilizers first and granular fertilizers second, in the hope of restoring the vegetative system of cocoa trees [11] [12] . Fertilizer appears to be one of the factors that could potentially revitalize the economy of an old cocoa-growing region. Research carried out in cocoa plantations in eastern Côte d’Ivoire has also shown that plots that received fertilizer had a significantly higher yield than the control plots that did not receive fertilizer [29] . However, differences in yield were observed between actual yield and potential yield. This difference could be linked either to black rot of cocoa pods caused by fungal agents of the genus Phytophthora, which severely infect the pods, or to damage caused by insects such as moths, which lay their eggs on the pods, causing them to rot. These yield differences may also be linked to rodents such as squirrels, which gnaw on pods and cocoa beans. Indeed, some research conducted on cocoa trees has shown that cocoa yield losses due to pod rot can range from 15.18% to 18.25% [30] .

As for the high number of wilted cherelles observed in cocoa farms, it should be stressed that the high rate of wilted cherelles strongly affects production. In fact, wilted cherelles act as a regulating factor for trees with a high number of pods, exceeding 100 pods per tree, which corresponds to around 4 tons of merchantable cocoa per hectare [31] ; other studies have shown that a rise in temperature increases the number of chérelle wilt [32] . The increase in flower cover, hence the appearance of cherelles, is thought to be due to the action of the phosphorus contained in the fertilizer. The results of several trials corroborate the major role of phosphorus in stimulating flowering [33] . The work of [28] on research stations has also shown the importance of phosphorus and potassium in improving cocoa yields. Phosphorus is considered the limiting factor for plants in tropical soils [34] .

In south-central Côte d’Ivoire, in the department of Divo, more specifically in Gazaville and Troucasso, the combined effect of certain doses of mineral and liquid fertilizers improved yield parameters. Treatments T4 (300 g NPK 0-23-19 + TE + 0.045 L NPK 11-9-41), T5 (300 g NPK 0-23-19 + TE + 0.09 L NPK 11-9-41) and T9 (150 g NPK 0-23-19 + TE + 0.18 L NPK 11-9-41) improved the number of wilted cherelles and yields at both sites. Treatments T4 and T5 had potential yields of 1886.62 kg∙ha⁻¹ and 2032.08 kg∙ha⁻¹ respectively at Troucasso. In Gazaville, on the other hand, treatments T4 (1920.37 kg∙ha⁻¹), T5 (2141.21 kg∙ha⁻¹) and T9 (1829.14 kg∙ha⁻¹) had the highest potential yields.

This study was conducted as part of a project initiated by the National Center for Agronomic Research (CNRA) and the Export Trading Company (ETG). The authors would like to thank CNRA and ETG for their financial and material support.