Abstract

Vietnam produces around 1.62 million tons of coffee per year, the second highest yield in the world after Brazil. Improving resource utilization efficiency has recently been identified as the major strategic goal of the industry. Appropriate mineral nutrition practices are pivotal to achieving this goal. The humid, tropical climate and acid soils of Vietnam create considerable challenges to achieving optimum balanced crop nutrition practices. The availability of alkaline elements, particularly potassium (K), calcium (Ca), and magnesium (Mg), is steadily declining. Polyhalite, a natural marine sedimentary mineral consisting of a hydrated sulfate of K, Ca, Mg and sulfur (S, 48%) was examined as a potential additive to composite NPK fertilizers. The mineral was tested in Kon Tum Province, as part of an alternative fertilization program for the coffee industry. A 3-year (2016-2018) experiment testing six fertilization practices was carried out in Dak Mar Commune at Dak Ha District on grey Feralic Acrisol. The region’s recommendation, farmers’ practice, consisted of 322, 82.5, and 270 kg ha^–1 N, P₂O₅, and K₂O (as MOP [muriate of potash]), respectively. The other five treatments had similar N and P₂O₅ rates of 100 and 667 kg ha^–1, but differed in K application rates and sources: zero-K control; MOP control with 120 kg K₂O ha^–1 K (received solely through MOP); mixed MOP and polyhalite (1:1), at rates of 120, 240, and 360 kg K₂O ha^–1. Leaf nutrient status exhibited significant fluctuations between pre- and post-application measurements. Almost all crop performance parameters were significantly improved in response to elevated K application rates, reaching a maximum at 240 kg K₂O ha^–1. Polyhalite application was at least equivalent to MOP in providing crop K requirements. Moreover, polyhalite improved trees’ Ca, Mg, and S status and stopped the decline of these nutrients in the soil. At equivalent K doses, polyhalite combined with MOP brought about higher coffee bean yields and quality, with consequently higher profits for farmers, when compared to MOP alone. In the long run, however, additional modifications to coffee fertilization practices should be examined in the region in order to stabilize tree nutrient status, thus supporting further enhancement of the region’s coffee industry.

Keywords: Balanced fertilization; calcium; Coffea robusta; mineral nutrition; Polysulphate; potassium; sulfur.

Introduction

Coffee is one of the leading agricultural commodities worldwide. According to FAO’s statistical data in 2018, total annual yield of world coffee product (green coffee) was 10.3 × 106 tons, 73.3% of which is produced by just six countries. Brazil led the list with 3.56 × 106 tons, followed by Vietnam, Indonesia, Columbia, Honduras, and Ethiopia, with 1.62; 0.42; 0.72; 0.48; and 0.47 × 106 tons, respectively (FAO, 2020). Vietnam accounts for about 16% of the world’s production. The steadily expanding plantation area, currently about 620,000 ha, means coffee is a major economic engine for the developing agricultural sector in Vietnam. The Central Highlands is the key region of coffee production in Vietnam, with nearly 90% of the country’s coffee area. More than half of this area, about 360,000 ha, is located in Dak Lak and Lam Dong. Therefore, much effort is made to improve the region’s coffee yield and quality.

Vietnam has unique achievements in developing Robusta coffee (Coffea robusta or Coffea canephora) as a high yielding cash crop. This has been made possible by intensification methods, including irrigation during the dry season (Marsh, 2007). Special concern has been devoted to nutrition requirements and fertilization dosage and regime. Coffee displays high demands for fertilizers, particularly nitrogen (N) and potassium (K) (Jessy, 2011); obtaining one ton of Robusta coffee beans would require 30-35 kg N, 5.2-6.0 kg phosphorus pentoxide (P₂O₅), 36.5-50.0 kg potassium oxide (K₂O); 4 kg calcium oxide (CaO); and 4 kg magnesium oxide (MgO), depending on tree age and soil types (Tiemann et al., 2018). A high producing coffee plantation would remove at least 135 kg N, 34 kg P₂O₅, and 145 kg K₂O ha^–1 (De Geus, 1973).

Potassium is needed for most basic processes in plants’ life cycle (Engels et al., 2012). In coffee, K requirements are high during the development of the berries and are at the maximum during their ripening. Peaks in rate of K uptake were observed immediately after bloom, prior to fruit ripening, and after harvest (Mitchell, 1988). Forestier (1969), studying Arabica coffee, showed that chronic lack of K brought about a significantly increased rate of young fruit abortion, degeneration of branches and consequent die-back.

Several studies on soil and fertilizer application for Robusta coffee in the commercial phase have been carried out, resulting in controversial information regarding the annual K requirements of coffee in Vietnam. Ton Nu Tuan Nam and Truong Hong (1993) concluded that reaching a maximum yield would require annual application of 250 kg K₂O ha^–1. Le Ngoc Bau (1997), who focused on plantations with particularly high coffee beans yields (>5 Mg ha^–1) in the Highland provinces (Gia Lai, Dak Lak and Kon Tum), found that the annual K dose ranged from 400-500 kg K₂O ha^–1, twice as much as recommended. Nevertheless, in cases where K was applied at levels 2-3 times higher than recommended, no significant effects on the yield or on the tree growth and development could be observed, and there was no correlation between yield and leaf K content. Nguyen Van Sanh (2009), who studied balanced fertilization in Ea Pok Coffee Cooperative (Dak Lak), showed that the appropriate K dose was 180 kg K₂O ha^–1. Truong Hong (1997) concluded that to furnish coffee bean yields higher than 2.6 Mg ha^–1, 250-260 kg K₂O ha^–1 was required on the basaltic soil of Buon Ma Thuot, compared to 125-180 kg K₂O ha^–1 for the gneiss soil in Kon Tum. A more recent study carried out in 2012-2014 at Dak Lak and Kom Tun Provinces, suggested an optimum application rate of 360 kg K₂O ha^–1 (Tien et al., 2015a). Displaying a very wide range, from 125-500 kg K₂O ha^–1, these studies fail to provide clear advice of the appropriate K application rate for Robusta coffee in the region.

Until the last decade, farmers in Vietnam tended to overuse fertilizers and did not consider the ratio among nutrition elements (Do Thi Nga, 2012). Excess doses of N and P fertilizers were usually applied, whereas K doses were very low. The imbalanced plant nutrition brought about relatively low resistance to pests, diseases, and other stress factors. The increasing awareness in Vietnam of the significance of K nutrition for coffee in recent years (Tien et al., 2015a) has given rise to considerable increases in the recommended K application rate; nevertheless, the optimum K application rate and practice are yet to be determined, requiring further research efforts.

The nature of the soil may be crucial to the quality and productivity. Soils in Vietnam developed from many different parental rocks including basalt, gneiss, granite, shale, limestone, lava and volcanic ash. Soil texture may vary from heavy loam to sandy soils with no obvious effects on coffee production as long as the soil layer is deep, easily drained but porous enough to hold considerable levels of water, air, and nutrients (Tiem, 1999). The humid tropic climate of Vietnam, however, creates significant challenges for soil nutrient availability. During the wet season, the liquid soil phase is prone to fluctuate very frequently, within hours or days, depending on the precipitation regime. This liquid phase contains most of the currently available nutrients, including K, that are leached away from the rhizosphere. In addition, the high precipitation rates intensify soil weathering and significantly increase soil acidity (Sanchez, 2019), which further reduces soil cation exchange capacity (CEC) and K availability (Zörb et al., 2014). Consequently, the opportunity window for plants to acquire K following application events is short and scarce, potentially leading to significant gaps between K application and uptake rates, considerable waste of fertilizer, and to environmental consequences.

Beyond K, the availability of alkaline elements – in particular calcium (Ca) and magnesium (Mg) – is steadily declining (Nam and Hong, 1993; Hong, 1997; Tien et al., 2015b). As with K, the tropical climate and frequent heavy precipitation accelerate Ca and Mg loss through leaching from the root zone. Consequently, deficiency symptoms often occur in plantations that were previously highly productive (Nguyen Van Sanh, 2009). Sulfur (S) is recognized as the fourth major plant nutrient after N, P, and K (Khan et al., 2005), and has been associated with high productivity (Zhao et al., 1999; Saito, 2004; Kovar and Grant, 2011). Sulfur often interacts with N to significantly enhance crop productivity (Jamal et al., 2010). However, current information regarding S application to acidic soils under tropical climates is scarce.

The recently introduced composite NPK fertilizers are not diverse enough to meet all nutrient requirements at each stage of growth, and on differing soils. Polyhalite is a natural mineral mined by Cleveland Potash, UK Ltd, and marketed as Polysulphate^® fertilizer. It occurs in sedimentary marine evaporates and consists of a hydrated sulfate of K, Ca, and Mg with the formula: K₂Ca₂Mg(SO₄)₄·2H₂O. The deposits found in Yorkshire, in the UK, typically consist of K₂O: 14%, SO₃: 48%, MgO: 6%, CaO: 17%. As a fertilizer providing four key plant nutrients – S, K, Mg, and Ca – polyhalite may offer attractive solutions to crop nutrition. In addition, polyhalite is less water soluble than more conventional sources (Barbarick, 1991; Yermiyahu et al., 2017; Yermiyahu et al., 2019) and is, therefore, a suitable fertilizer to supply these four nutrients during the rainy growing season. Once a proper application is established, polyhalite may not only provide a significant part of crop K requirements, but also supply secondary macronutrients that are essential under the present cropping environment of Robusta coffee in Vietnam.

The objectives of the present study were to determine an optimum K application rate for the Central Highlands region, evaluate and compare the effects of combining two K sources, muriate of potash (MOP [potassium chloride, KCl]) and polyhalite, on performance, yield, and quality, and consequently, determine the agronomic and economic efficiencies of the fertilizer combinations in Robusta coffee production.

<h3>Results</h3>
<h4>Leaf nutrient status</h4>
<p>Leaf nutrient status, related to N, P, K, S, Ca, and Mg, before fertilizer applications was suboptimal throughout the 3-year experiment (Fig.&nbsp;1). The response of leaf nutrient status to fertilizer application events was almost immediate; 20 days after application of a particular nutrient, its concentration in the indicative leaves returned to the recommended optimum range, usually corresponding to the application rate (Fig.&nbsp;1).</p>

Leaf N concentration fluctuated from about 2.6% before each urea application, below the recommended threshold, to 3.4%, within the recommended optimum, after it (Fig. 1A), displaying considerable instability. Leaf P concentration, which varied from 0.09-0.1% before fertilizer application, rose to the recommended range from 0.12-0.13% (Fig. 1B). Before fertilizer application, leaf K concentration ranged from 1.92-1.95%, slightly below the minimum threshold (Fig. 1C). After application, however, it tightly corresponded with K application rate remaining very low (1.82%) at K₀ (where no K was applied), highest (2.21 and 2.25%) at FP (270 kg K₂O ha^–1 year^–1) and at MOP:polyhalite K_180:180 (360 kg K₂O ha^–1 year^–1), and moderate (2.12%) but within the recommended optimum where applied with 120 kg K₂O ha^–1 year^–1 (Fig. 1C).

Leaf S concentration ranged from 0.1-0.11% before fertilizer application events. Interestingly, leaf S concentration exhibited a slight but consistent rise following fertilizer application, even when the application did not contain the nutrient. However, the leaf S concentration response to polyhalite application rate was linearly positive (Fig. 1D). In the control treatments (FP K₂₇₀, K₀, and MOP K₁₂₀), leaf Ca and Mg concentrations tended to slightly but consistently decrease in response to fertilizer application events, remaining well below the recommended range. However, similar to the case of S, polyhalite application brought about considerable increases in leaf Ca and Mg status, which tightly corresponded with the fertilizer application rate (Fig. 1E, F).

Vegetative growth

The vegetative growth and development of the coffee trees were significantly affected by the fertilizer treatments (Fig. 2). Branch elongation and the number of internodes developed during the growing season were both lesser at the K₀ treatment, where K was not included in the fertilization practice. Potassium application rates greater than 120 kg K₂O ha^–1, supplied through polyhalite or MOP, brought about the highest vegetative growth, significantly greater than at K₀. The lower K rate (120 kg K₂O ha^–1) resulted in intermediate growth levels with no significant differences between trees applied with either MOP or polyhalite + MOP (Fig. 2).

Susceptibility to diseases

The infection rates of coffee rust (Hemileia vastatrix) observed during the 3-year experiment were quite low (3.25-4.6%), with no significant influence from the fertilizer treatments (Fig. 3A). In contrast, considerably higher infection rates of green scale (Coccos viridis) were observed, with 4.5-13% of the trees infected. The infection rates were significantly higher at treatment K₀, and progressively declined with the increasing K application rate, displaying a significantly lower value at the MOP:polyhalite K_180:180 treatment (Fig. 3B).

Yield indicators

Pre-harvest fruit shedding was highest, 38%, at the K₀ treatment (Fig. 4A). MOP K₁₂₀ treatment significantly reduced fruit shedding to about 30%; however, fruit shedding was further alleviated at higher K application rates, with shedding rates below 25% when polyhalite was included (Fig. 4A).

Fresh fruit weight (Fig. 4B), as well as fruit size (Fig. 4C), were remarkably affected by the fertilizer treatments. Both parameters exhibited significantly low values at the K₀ treatment and showed clear positive responses to the rise in K application rates. The MOP:polyhalite K_60:60 gave rise to a higher fruit weight compared to the corresponding MOP K₁₂₀ treatment (Fig. 4B). In addition, MOP:polyhalite K_120:120, with a rate of 240 kg K₂O ha^–1, yielded higher fruit weight and size compared to FP, with 270 kg K₂O ha^–1 (Fig. 4B, C). Nevertheless, an additional increase of K rate to 360 kg K₂O ha^–1 (MOP:polyhalite K_180:180) did not result in a further increase in fruit weight or size.

Fruit/bean ratio was significantly higher (5.60) at K₀, decreased to an intermediate range (4.45-4.60) at rates of 120 and 270 kg K₂O ha^–1, but was significantly lower (4.37 and 4.33) under the MOP:polyhalite mixtures at 240 and 360 kg K₂O ha^–1, respectively (Fig. 4D).

Fruit and bean yields

The response patterns of both whole fruit and coffee bean yields to the various fertilizer treatments were similar (Fig. 5A, B). Yields were extremely low at K₀ and significantly increased in response to K application at rates ranging from 120-270 kg K₂O ha^–1. Maximum yields were obtained at 240 kg K₂O ha^–1 of mixed MOP and polyhalite, with no further response to the highest K rate (Fig. 5A, B). The response of bean yield to K application rate obeys a quadratic function, where the maximum yield is reached at about 300 kg K₂O ha^–1, and begins to decline with further increases in K rate (Fig. 5C). The agronomic K efficiency declines sharply from 15 to 6 kg beans kg^–1 K₂O as K rate increases up to 360 kg K₂O ha^–1 (Fig. 5D). Interestingly, both parameters were significantly higher for the combined MOP and polyhalite compared to MOP, at K rates between 120-270 kg K₂O ha^–1.

Coffee bean quality

Bean size, an important quality measure in the coffee industry, was evaluated using two parameters: bean weight, and the fraction of large beans (length greater than 6.3 mm). Bean weight, which was very low (8 g 100^–1beans) at K₀, significantly increased in response to the stepwise rise in K application rate and was almost double at 240 kg K₂O ha^–1 (Fig. 6A). Interestingly, bean weight response to the mixed fertilizer (MOP and polyhalite) treatments was significantly stronger than the equal MOP K rate application (120 kg K₂O ha^–1). Furthermore, bean weight was slightly greater at the MOP:polyhalite K_120:120 (240 kg K₂O ha^–1) compared to that of the higher MOP K rate (270 kg K₂O ha^–1) at the FP (K₂₇₀). Nevertheless, no further response of bean weight was observed when the K rate was raised to 360 kg K₂O ha^–1 (Fig. 6A).

The response patterns of bean weight to the fertilizer treatments corresponded closely with the fraction size of large beans (> 6.3 mm); however, the differences between the mixed MOP and polyhalite and the pure MOP treatments were less pronounced (Fig. 6B).

The concentrations of caffeine, chlorogenic acid, and trigonelline, highly desirable natural compounds and important indicators of coffee quality, displayed considerable responses to the fertilizer treatments (Fig. 7). Caffeine concentration showed a significant linear increase as a function of K application rate (Fig. 7A). Chlorogenic acid and trigonelline displayed similar response patterns; however, the response rates slightly decreased with the rising K rates, and hence, a more accurate description was provided using quadratic rather than linear functions (Fig. 7B, C). No consistent effects were observed for polyhalite with regard to these coffee quality indicators.

Economic considerations

Potassium application, with or without polyhalite, brought about a relatively small increase in the cost of coffee cultivation, ranging from 11-28 million VND ha^–1 year^–1, compared to no K application. The difference in cost between the local farmers’ fertilization practice and the highest fertilizer rates in the experiment were much smaller – less than 12 million VND ha^–1 year^–1 (Fig. 8). In contrast, K application rates ranging from 120-360 kg K₂O ha^–1 gave rise to 72-123% increase in income compared to no K application (Fig. 8). The boost in profit was especially pronounced, increasing from 10.4 at K₀ to 50-70 million VND ha^–1 year^–1 where K fertilization was practiced. The highest profit, 69.8 million VND ha^–1 year^–1, was obtained at the MOP:polyhalite K_120:120 treatment, 9.2 million VND ha^–1 year^–1 (15.3%) greater than that of the farmers’ practice. Additionally, an increase in K application rate to 360 kg K₂O ha^–1 (MOP:polyhalite K_180:180) did not yield any further effect of the profit. At the lower K rates, 120 kg K₂O ha^–1, the combined MOP and polyhalite treatment had a slightly higher profit compared to MOP alone, 55 and 50.2 million VND ha^–1 year^–1, respectively, but less than that of the farmers’ practice (Fig. 8).

Effects on soil fertility

Soil properties and nutrient status were evaluated in 2016, at the beginning, and after three years at the end of the experiment. Soil acidity (pHKCl of 4.0-4.1) and organic matter content (2.85%) did not change throughout the 3-year experiment. Total N, P₂O₅, and K₂O were poor at the beginning of the experiment, with 0.17, 0.08, and 0.11%, respectively. Soil K content extensively declined from 97.4 to 90.5 mg K₂O kg^–1 at the K₀ treatment. Much smaller decreases were observed at K application levels of 120 kg K₂O ha^–1, whereas soil K levels slightly and proportionally increased at rates of 240 kg K₂O ha^–1 and above (Fig. 9A).

Soil S content slightly decreased but was maintained at a range from 20-21 mg kg^–1 at the three treatments that were not applied with polyhalite. However, soil S considerably increased in proportion to the polyhalite application rate (Fig. 9B). Similarly, soil available Ca content tended to slightly drop in the absence of Ca application, but clearly and proportionally increased after the polyhalite application during the 3-year experiment (Fig. 9C). A similar pattern was observed for soil available Mg, which decreased in the absence of Mg application, and increased with the polyhalite treatments (Fig. 9D).

Discussion

None of the fertilizer treatments had a long-term effect on soil acidity, which remained very high (pHKCl 4.0-4.1). High soil acidity, a serious problem typical of humid tropical regions (Mintesnot et al., 2015; Tiemann et al., 2018; Byrareddy et al., 2019; Sanchez, 2019), requires systematic solutions that are beyond the scope of the present study. The significant degradation of soil available K in the absence of K application (K₀) indicates that the intrinsic soil sources for this nutrient are poor and cannot support long-term coffee production. Soil K recovery was slightly improved at 120 kg K₂O ha^–1, however, a tendency for soil K buildup was noticed only at an application rate of 240 kg K₂O ha^–1 and above (Fig. 9A). Polyhalite demonstrated a capacity to support soil K status at a level which was at least similar to that of MOP. In this sense, however, polyhalite application halted the degradation of soil available Ca, Mg, and S, demonstrating considerable improvement in soil nutrient status, from poor soil S, and very poor soil Ca and Mg at the beginning of the experiment to more acceptable levels at its end (Fig. 9). Consecutive polyhalite applications, therefore, may enrich the soil with these nutrients, among which Ca has significant effects on soil structure and texture (Sanchez, 2019).

Nevertheless, the substantial fluctuations in the leaf nutrient concentrations, before and after fertilizer applications (Fig. 1), indicate a chronic inability of productive coffee trees to rely on the fertility of the local soil. Moreover, this phenomenon highlights the fragile nutrient supply under the current fertilizer practices. Although the duration of maintained optimal leaf nutrient status between applications remains obscure, the fluctuations during the growing season were significant even under the high application rates suggesting a need for further efforts to stabilize the mineral nutrition of coffee trees in the region.

Although not within the focus of the present study, N may serve as a suitable example for such needs. Nitrogen was applied in the form of urea, a fertilizer known for providing very high N availability to plants, but also for having a very short durability in the soil (Witte, 2011). Nitrogen use efficiency in crop production is generally well below 50%, resulting in economic losses and creating ecological problems including groundwater pollution and emission of nitric oxides (Ladha et al., 2005; Yadav et al., 2017). The seasonal N dose in the experiment was 300-322 kg ha^–1, which might be suitable for coffee production on the Feralic Acrisol of Dak Ha, Kon Tum province. However, N fertilization practice for coffee in the region seems to require reassessment in order to optimize tree nutrition, increase the farmers’ profits, and avoid negative environmental consequences.

The tight relationship between nutrient availability and leaf nutrient concentration was demonstrated for K, Ca, Mg, and S (Fig. 1); however, stabilizing nutrient flux and subsequently leaf nutrient concentration during the growing season is essential in the efforts to enhance coffee yields and quality. Nitrogen appears to be very important in this regard; it seems that once N is applied, the uptake and temporary leaf concentration of all the other nutrients examined increase, depending on their availability (Fig. 1). Sooner or later, when N concentration diminishes, the other nutrients decline as well. The pivotal role of N in plant metabolism and growth, and the strong interactions it has with other nutrients (Jamal et al., 2010; Marschner, 2012) provide a reasonable explanation for the leaf nutrient fluctuations.

In addition to the need for stabilizing nutrient uptake during the growing season, it is necessary to increase fertilizer supply during the dry season, before plant growth and reproductive development restart (Tiemann et al., 2018). In fact, well-furnished tree nutrient reserves at the beginning of the growing season were shown to significantly enhance coffee crop performance and yield (Forestier, 1969; Lima-Filho and Malavolta, 2003; Malavolta, 2003). With the threatening effects of climate change on the coffee industry in the tropics (Thioune et al., 2020), supplementary irrigation during the dry season should be revisited (D’haeze et al., 2017; Byrareddy et al., 2019) and furthermore, be used for nutrient supply through fertigation.

In an era of climate change and the rising risk of transient water-stress periods, the significance of adequate K supply is highlighted. Among many other important roles in plant physiology, K is pivotal to plant-water relations and management (Marschner, 2012; Zörb et al., 2014). Unequivocally, K application rates ranging from 120-270 kg K₂O ha^–1 brought about tremendous enhancements in all aspects of coffee crop performance, including vegetative growth (Fig. 2), pest tolerance (Fig. 3B), fruitlet persistence and yield determinants (Fig. 4), fresh fruit and bean yield parameters (Fig. 5), and coffee bean quality traits (Fig. 6). In most cases, no advantages could be identified for any further increase in K rates. These results suggest a slight reduction of the recommended K dose, particularly if it would be partially supplied through polyhalite.

The effects of polyhalite on coffee crop performance were less significant compared to those of the K rates, which might be attributed to the present study’s three year duration being inadequate for long-term evaluations. Nevertheless, polyhalite demonstrated reasonable capacity as a K donor fertilizer, which was even better than MOP in several aspects such as fresh fruit weight (Fig. 4B), bean weight (Fig. 6A), and agronomic K efficiency (Fig. 5D). However, the exact reasons for the differences between MOP alone and MOP + polyhalite treatments require further research, as there was only one case where K rates were similar (MOP K₁₂₀ and MOP:polyhalite K_60:60). Clearly, polyhalite application enhanced the crop nutrient status, increasing leaf Ca, Mg, and S concentrations (Fig. 1).

Most of the recent analyses of the Vietnamese coffee industry point to the essential need to optimize crop fertilizer practices through a more balanced macro and micronutrient application in order to increase coffee yield and quality, and to avoid nutrient pollution of soil and water resources due to excess spreading of fertilizer (De Geus, 1973; Tiemann et al., 2018; Byrareddy et al., 2019). Interestingly, the current farmers’ practice in the region meets N, P, and K requirements, but it fails to supply essential macronutrients such as Ca, Mg, and S, as evidenced by leaf nutrient status (Fig. 1D-F) and by the degrading soil status of these nutrients (Fig. 9B-D). Polyhalite application appears to pave a way toward addressing imbalanced nutrition and introducing nutrients into the coffee production system that are currently lacking. An economic analysis shows that polyhalite application also helps farmers’ to increase their profits. At an application rate of 240 kg K₂O ha^–1, replacing 50% of dose with polyhalite gave rise to much greater profits, compared with equivalent or even higher MOP application rates (FP K₂₇₀) (Fig. 8C).

Conclusions

Balanced mineral nutrition is essential for maintaining and enhancing coffee production in the Central Highlands of Vietnam. Without balanced nutrition, coffee yield and quality, as well as soil fertility, would continue to degrade. Polyhalite application was at least equivalent to MOP in providing crop K requirements. Moreover, polyhalite improved trees’ Ca, Mg, and S status, while also stopping the decline of these nutrients in the soil. At equivalent K doses, polyhalite mixed with MOP brought about higher coffee bean yields and quality, with consequently higher profits to farmers, compared to MOP alone. In the long run, however, additional modifications of coffee fertilization practices should be examined in the region in order to stabilize tree nutrient status, thus supporting further enhancement of the region’s coffee industry.

Acknowledgements

We thank Dr. Nguyen Van Bo and Mr. Gershon Kalyan for giving their advice and assistance for the study, and we thank Mrs Hoang Le Thuy and staff at the Central Highland Soils and Fertilizers Research Centre for taking care the field experiment. This study was funded by the International Potash Institute via the research project “Investigation of the Agronomic Efficiency of Polysulphate on Yield and Quality of Some Crops in Vietnam”.

References

• Barbarick, K.A., 1991. Polyhalite Application to Sorghum-Sudangrass and Leaching in Soil Columns. Soil Science 151(2):159-166.
• Bray, R.H. and L.T. Kurtz. 1945. Determination of Total, Organic, and Available Forms of Phosphorus in Soils. Soil Sci., 59:39–45.
• Byrareddy, V., L. Kouadio, S. Mushtaq, and R. Stone. 2019. Sustainable Production of Robusta Coffee under a Changing Climate: a 10-year Monitoring of Fertilizer Management in Coffee Farms in Vietnam and Indonesia. Agronomy, 9(9):499.
• Chesnin, L. and C.H. Yien. 1951. Turbidimetric Determination of Available Sulfates 1. Soil Science Society of America Journal, 15(C):149-151.
• De Geus, J.G. 1973. Fertilizer Guide for Tropicals and Subtropicals. 2^nd edition, Centre d’Etude de I’Azote Zurich. p. 440-471.
• D’haeze, D., P. Baker, and P. Van Tan. 2017. Vietnam’s Central Highlands’ Upland Agriculture under Pressure Because of the Looming Effects of Climate Change—Focus on Robusta Coffee. In: Proceedings of the Conference: Buon Ma Thout coffee festival, Buon Ma Thout, Vietnam, 9–16 March 2017.
• Do Thi Nga. 2012. Research on the Competitive Ability of Coffee Products in Different Business Sectors in Dak Lak Province. PhD thesis, Hanoi Agricultural University, Hanoi.
• Engels, C., E.A. Kirkby, and P. White. 2012. Mineral Nutrition, Yield and Source-Sink Relationships. In: Marschner P. (ed.). Marschner’s Mineral Nutrition of Higher Plants, 3^rd edition. Elsevier Ltd. p. 85-134.
• FAO, 2020. http://www.fao.org/faostat/en/#data/QC
• Forestier, F. 1969. New Problems Used Mineral Fertilizer on Coffee in Republic of Middle Africa. The Café - Cacao 1/1969.
• Jamal, A., Y-S., Moon, and M.Z. Abdin. 2010. Sulphur - A General Overview and Interaction with Nitrogen. Australian J. Crop Sci. 4:523-529.
• Jessy, M.D. 2011. Potassium Management in Plantation Crops with Special Reference to Tea, Coffee, and Rubber. Karnataka J. Agric. Sci. 24(1):67-74.
• Khan, N.A., M. Mobin, and Samiullah. 2005. The Influence of Gibberellic Acid and Sulfur Fertilization Rate on Growth and S-Use Efficiency of Mustard (Brassica juncea). Plant and Soil 270:269-274.
• Kovar, J.L. and C.A. Grant. 2011. Nutrient Cycling in Soils: Sulfur. Publications from USDA-ARS/UNL Faculty. Paper 1383. http://digitalcommons.unl.edu/usdaarsfacpub/1383.
• Ladha, J.K., H. Pathak, T.J. Krupnik, J. Six, and C. van Kessel. 2005. Efficiency of Fertilizer Nitrogen in Cereal Production: Retrospects and Prospects. Advances in agronomy 87:85-156.
• Le Ngoc Bau. 1997. Investigating the Technologies for Improving Robusta Coffee in Dak Lak Province. MSc thesis, Hanoi Agricultural University, Hanoi.
• Lima Filho, O.D. and E. Malavolta. 2003. Studies on Mineral Nutrition of the Coffee Plant (Coffea arabica L. cv. Catuaí Vermelho): LXIV. Remobilization and Re-Utilization of Nitrogen and Potassium by Normal and Deficient Plants. Brazilian journal of biology, 63(3):481-490.
• Malavolta, E. 2003. The Mineral Nutrition of Coffee. Studies on the Mineral Nutrition of the Coffee Plants. Paper No. 46, Notes: H 8.1.5.1 #4251.
• Marschner, P. 2012. Marschner’s Mineral Nutrition of Higher Plants, 3^rd edition. Elsevier Ltd.
• Marsh, A. 2007. Diversification by Smallholder Farmers: Viet Nam Robusta Coffee. Agricultural Management, Marketing and Finance Working Document 19. FAO, Rome.
• Mintesnot, A., N. Dechassa, and A. Mohammed. 2015. Association of Arabica Coffee Quality Attributes with Selected Soil Chemical Properties. East African Journal of Sciences, 9(2):73-84.
• Mitchell, H.W. 1988. Cultivation and Harvesting of the Arabica Coffee Tree. In: Clarke, R.J., and R. Macre (eds.). Coffee. Agronomy, Elsevier Applied Science, London 4(2):43-90.
• Nam, T.N.T. and T. Hong. 1993. Research Results of Applying NPK Compound Fertilizers for Robusta Coffee on Two Sites of Basaltic Soil in Dak Lak Province. Scientific Report for Ministry of Agriculture and Rural Development.
• Nguyen Van Sanh. 2009. Research on Nutrient Deficiency Diagnostic in Coffee Leaf and its Application for Fertilizer Recommendation for Robusta Coffee in Dak Lak Province. PhD thesis, Hanoi Agricultural University, Hanoi.
• Saito, K. 2004. Sulfur Assimilatory Metabolism. The Long and Smelling Road. Plant Physiol. 136:2443-2450.
• Sanchez, P.A. 2019. Properties and Management of Soils in the Tropics. Cambridge University Press, Cambridge, UK, 2019.
• Tiem, H.T. 1999. The Vietnamese Coffee. Agricultural Publishing House, Hanoi.
• Tiemann, T., T.M. Aye, N.D. Dung, T.M. Tien, M. Fisher, E.N. de Paulo, and T. Oberthür. 2018. Crop Nutrition for Vietnamese Robusta Coffee. Better Crops with Plant Food 102:20-23.
• Tien, T.M., H.C. Truc, and N.V. Bo. 2015a. Effects of Annual Potassium Dosage on the Yield and Quality of Coffea robusta in Vietnam. IPI e-ifc 41:13-20.
• Tien, T.M., H.C. Truc, and N.V. Bo. 2015b. Potassium Application and Uptake in Coffee (Coffea robusta) plantations in Vietnam. IPI e-ifc 42:3-9.
• Thioune, E.H., S. Strickler, T. Gallagher, A. Charpagne, P. Decombes, B. Osborne, and J. McCarthy. 2020. Temperature Impacts the Response of Coffea canephora to Decreasing Soil Water Availability. Tropical Plant Biology 13:236–250.
• Ton Nu Tuan Nam and Truong Hong. 1993. Research Results of Applying NPK Compound Fertilizers for Robusta Coffee on Two Sites of Basaltic Soil in Dak Lak Province. Scientific Report for Ministry of Agriculture and Rural Development.
• Truong Hong. 1997. Determining Suitable NPK Compound Fertilizers for Robusta Coffee on Reddish Brown Basaltic Soil in Dak Lak Province and Grey Granite Soil in Kon Tum Province. PhD thesis, Institute of Agricultural Science for Southern Vietnam, Ho Chi Minh City.
• Walkley A. 1947. A Critical Examination of a Rapid Method for Determination of Organic Carbon in Soils – Effect of Variations in Digestion Conditions and of Inorganic Soil Constituents. Soil Sci. 63:251–257.
• Walsh, A. 1955. The Application of Atomic Absorption Spectra to Clinical Analysis. Spectrochim. Acta 7:108.
• Witte, C.P. 2011. Urea Metabolism in Plants. Plant Science 180(3):431-438.
• Yadav, M.R., R. Kumar, C.M. Parihar, R.K. Yadav, S.L. Jat, H. Ram, and A. Ghosh. 2017. Strategies for Improving Nitrogen Use Efficiency: A Review. Agricultural Reviews 38(1):29-40.
• Yermiyahu, U., I. Zipori, I. Faingold, L. Yusopov, N. Faust, and A. Bar-Tal. 2017. Polyhalite as a Multi Nutrient Fertilizer – Potassium, Magnesium, Calcium and Sulfate. Israel Journal of Plant Sciences 64(3-4):145-157.
• Yermiyahu, U., I. Zipori, C. Omer, and Y. Beer, 2019. Solubility of Granular Polyhalite under Laboratory and Field Conditions. International Potash Institute (IPI) e-ifc 58:3-9.
• Zhao, F.J., M.J. Hawkesford, and S.P. McGrath. 1999. Sulfur Assimilation and Effects on Yield and Quality of Wheat. J. Cereal Sci. 30:1-17.
• Zörb, C., M. Senbayram, and E. Peiter. 2014. Potassium in Agriculture–Status and Perspectives. Journal of Plant Physiology 171(9):656-669.