INTRODUCTION

Sweet potato has been classified under the nomenclature Ipomoea batatas (L.) Lam. by Lamarck and Poiret (1791). It is an herbaceous, creeping, perennial plant used as an annual crop because of its tuberous roots, which can vary in shape and flesh color (Florez et al., 2016). Its origin is American, and its evolution may have occurred separately in Central and South America (Venezuela, Colombia, Ecuador, and Peru); however, the origins of its domestication are unclear (Morales et al., 2017;Roullier et al., 2013). Documents such as those present in the description of the “Encomienda en Paraguay” on the establishment of this crop in Paraguay point to its pre-Columbian presence in the country (Elman, 1951;Enciso and Gonzales, 2023).

The adaptive developmental characteristics of the crop and its nutritional properties make sweet potatoes relevant to many tropical and subtropical countries (Otálora et al., 2024). It is a famine food whose roots provide significant amounts of carbohydrates, beta-carotene, vitamin C, vitamin B complex, and E, while the leaves contain high amounts of polyphenols, anthocyanins, and proteins (Lebot, 2009;Lee et al., 2007;Woolfe, 1992).

Despite its relevance and outstanding production potential, sweet potato production has declined worldwide (FAOSTAT, 2024). According to Kays (2005), the problem lies in its limited scope of use and acceptance, since it is mostly used unprocessed, for direct consumption. The report emphasizes that plant breeding is the most convenient way to address the deficiencies that modulate its utilization and consumption. On the other hand, Brandenberg et al. (2014) mention that the main challenges are obtaining investment, proper crop management, and profitable practices that include certified seeds and good production practices. In Paraguay, sweet potato production is mainly carried out by small-scale farmers (DCEA-MAG, 2023), who face difficulties such as a lack of critical assets, technological know-how, fragile infrastructure, and limited coordination regarding farmers’ associations and production and selling timing (FAO, 2021).

The main producer of sweet potatoes in the world is China, with an annual average of almost 50 million tons between 2018 and 2022, representing 56% of the world production, according to data reported in the database and statistics of the Food and Agriculture Organization of the United Nations (FAOSTAT, 2024). Data from the National Agricultural Census (CAN), conducted by the Department of Census and Statistics of the Ministry of Agriculture and Livestock of Paraguay in 2022 shows that the national production of sweet potatoes is 46,836 tons with an average yield of 5.62 ton/ha (DCEA-MAG, 2023). Compared to countries in the South American region, the national sweet potato yields of Paraguay are lower. Argentina has a yield of 15 ton/ha (IDEP, 2020), and Brazil, the main importer of sweet potatoes to Paraguay, has a national average of 30 ton/ha (Cavalcante et al., 2021). According to the SICEX records from the Central Bank of Paraguay, the average sweet potato importation from Brazil per year between 2020 and 2023 was 4,063.25 tons (BCP, 2024), evidencing a deficit in the national supply of sweet potatoes compared to the demand for the domestic market.

This research was conducted jointly within the framework of the KOPIA project and IPTA, with the objectives of selecting sweet potato varieties adapted to the local climate and the national market preferences and improving crop yields through developing high-yielding cultivation technologies.

MATERIALS AND METHODS

Variety selection: For the selection of the varieties with the best productivity performance, evaluations of eleven sweet potato varieties were carried out in two of IPTA’s regional institutes: Centro de Investigación para la Agricultura Familiar in Choré (IPTA-Choré), department of San Pedro, and Campo Experimental de San Juan Bautista in the department of Misiones (IPTA-San Juan Bautista). At the region of IPTA-Choré, the average annual temperature is 23°C, and the average annual precipitation is 1,551 mm according to the meteorological station at the Choré Experimental Field. The execution period of the experiments at IPTA-Choré was between 2021/2022 and 2022/ 2023. In the first period, the soil pH ranged from 5.5 to 5.8 and was considered of medium fertility. In the second period, the soil pH was below 5 and was considered low fertility, to replicate the soil conditions usually found in the farmers’ plot. At the IPTA-San Juan Bautista region, the average annual temperature is 22°C and the average annual precipitation is 1,600 mm according to the meteorological station of San Juan Bautista. Here, the experiment was carried out only in the period 2021/2022, and the soil was considered low to medium fertility, according to the national fertility classification (Fatecha et al., 2017). For the selection of varieties, the following cultivars were subjected to yield measurement in ton/ha: Jety Paraguay, Pyta Guazu, Jety Uruguayo, Taiwanés, Japonesa I, Japonesa II, Princesa, Roxa, Chaco I, Chaco II, and Andaí. Ridges with drip irrigation systems were used, however, no fertilization was applied. The planting distance was 0.8 m between rows and 0.35 m between plants. The experimental design was a randomized block with 11 treatments consisting of different varieties and 3 replications, totaling 33 experimental units. Each experimental unit consisted of 4 rows, each 5 m long and 3.2 m wide, with 0.8 m spacing between them, totaling 16 m² per experimental unit. In the first stage of the crop (15 days from planting to emergence), the irrigation frequency was three times per week; in the second, twice per week; and in the last (flowering stage), it was according to the crop’s necessity. The harvest of all varieties was conducted 120 DPP, using the two central rows of each experimental unit and leaving 0.5 m at each corner of each central row. The results were analyzed using the Tukey variance test at 5%.

Alternative planting and harvesting periods: The experiments were conducted at IPTA-Choré in the department of San Pedro at different times between 2022 and 2023. A total of eight different planting periods were used with different harvesting times between 120 and 135 DPP. The varieties used were Andaí, Jety Paraguay, and Pyta Guazu, which were planted in ridges. Each experimental unit consisted of 4 rows 5 m long and spaced 0.8 m apart, totaling 16 m² of surface area per experimental unit. Besides natural precipitation, no additional fertilization was implemented. The yield was evaluated in ton/ha for each period and the results were analyzed using the Tukey variance test at 5%.

Use of ridges: The evaluation of different planting methods with and without ridges was conducted at the IPTA-Choré Research Center in the 2021/2022 period. Three varieties were used: Andaí, Jety Paraguay, and Pyta Guasu, and three planting methods using cuttings: spiral, vertical, and curved. These planting methods were chosen based on the common practice observed among farmers in Paraguay. However, the spiral method is mainly used by the indigenous of the Chaco region, where they harvest the tubers as they need all year round. The cutting length for the vertical and curved methods was 25 to 30 cm and 80 cm for the spiral (Fig. 5). Regarding nodes, the vertical cuttings had about 3, the curved about 7, and the spiral about 10 or more. The yield and its increase were evaluated in ton/ha, comparing plantations using ridges and others without them. In both groups (experimental and control), no additional or controlled irrigation systems were implemented besides natural precipitation, and no fertilizers or phytosanitary treatments were used. The sweet potato was harvested 120 DPP for all varieties used. The experiment was conducted in a randomized block design with 2 main treatments (with and without ridges), 4 secondary treatments (different planting methods), and 3 replications. Each experimental unit consisted of 4 rows of 3.2 m wide, 5 m long, and 0.80 m spacing between them, a cultivation area of 16 m². The total experiment area comprises 24 experimental units of 16 m². The results were analyzed using the Tukey variance test at 5%.

Fertilizer determination: The determination and adequacy of chemical fertilizer doses were carried out in two crop planting periods: March 2023 and September 2023, at the Colegio Técnico Agropecuario Kamba Rembe, in the district of San Vicente Pancholo, department of San Pedro. The variety used was Chaco I, in ridges and curved planting method. The yield was evaluated in ton/ha for each treatment with increasing doses of nitrogen, phosphorus, and potassium. As the nitrogen source, urea was used with 45% N; as the phosphorus source, triple superphosphate was used with 46% P₂O₅; and as the potassium source, potassium chloride was used with 60% K₂O. For the control group, no fertilizer was used. Each treatment consisted of different levels of N, P, and K fertilization, as detailed in Table 1. Drip irrigation systems installed flush with the ridges were used. In the first stage of the crop (15 days from planting to emergence), the irrigation frequency was three times per week; in the second, twice per week; and in the last (flowering stage), it was according to the crop’s necessity. The different doses were applied directly in rows on the ridges before planting. An incomplete 3x4 factorial design was used with an additional plot arranged in a randomized block. The first factor was the N, P, and K, macronutrients and the second factor was four levels of fertilization. An additional plot without fertilization was installed for the control. Each experimental unit consisted of 4 rows of 5 m long, 0.8 m spacing between rows, and 0.35 m distance between plants, totaling 16 m². Three replications were carried out, reaching 33 experimental units. All results were analyzed using the Tukey variance test at 5%.

RESULTS AND DISCUSSION

Variety selection: Six varieties were selected for their performance in productivity and quality: Andaí, Jety Paraguay, Pyta Guasu, Jety Uruguayo, Chaco I, and Taiwanes. Among the varieties evaluated at IPTA-Choré, the statistical results of the experiments carried out in the 2021/ 2022 and 2022/2023 periods showed significant differences between varieties, as shown in Table 2. Andaí obtained the highest yield in both periods, with the best being 34.7 ton/ha in 2021/2022. Overall, productivity was higher in 2021/2022 for all varieties tested, considering the soil fertility was higher than in the second period. However, similar yield patterns were observed across both periods for each variety when compared to others (Fig. 1).

In the evaluation of the varieties in IPTA-San Juan Bautista, Jety Uruguayo obtained the highest yield at 23.6 ton/ha; followed by Chaco I and Taiwanes (Table 2). Andaí, Jety Paraguay, and Chaco I were chosen to install the demonstration seedbeds, being Chaco I the variety with the best adaptation to the commercial demands of the growers. The average yield results of all selected varieties across IPTA-Choré and IPTA-San Juan Bautista showed an increase over the 2022 national average yield of 5.65 ton/ha, according to the CAN (DCEA - MAG, 2023).

Comparing sweet potato performance across different locations, Enciso et al. (2021) assessed five varieties (Ib-023, Morotí, Morada INTA, Morado, and Sa’yju) in two zones with different soil and climate conditions (Loma Plata and San Lorenzo). Morotí, Ib-023, and Morada demonstrated superior total yield in the Loma Plata area compared to San Lorenzo. In contrast, Morada INTA did not show statistically significant differences between the two locations, while Sa’yju exhibited a higher yield exclusively in San Lorenzo. This underscores the significant interaction between locations and genotypes in root yield, highlighting the crucial role of genotype selection for specific localities.

Furthermore, Mbwaga et al. (2007) and Brandenberger et al. (2014) affirm that genotype-environment interactions strongly influence crop yield, with a variety performing differently across varied soil and climate conditions. Moreover, Yahaya et al. (2015) and Islam et al. (2002) note significant correlations between leaf number per plant, average root weight, number of roots per plant, and yield, emphasizing that average root weight directly contributes significantly to total crop yield. Enyi (1977) also indicates a negative correlation between the percentage of total dry matter allocated to vegetative parts and tuber yield.

Alternative planting and harvesting periods: The results show that sweet potato planting time has a significant influence on productivity. Statistically, the best planting period evaluated was December. However, there is no significant difference when compared to September and June. The optimal planting period varies depending on the variety used. For the Andai and Jety Paraguay varieties, the ideal planting time is in December, with a harvest at 122 days post-planting (DPP), yielding 25.86 and 22.48 ton/ha, respectively. In contrast, the Pyta Guasu variety achieves its highest yield when planted between September and October, with harvests at 122 and 124 DPP, producing 15.99 and 17.66 tons per hectare, respectively. Table 3 and Fig. 2 illustrate the productivity variations for each planting period across the three different varieties.

In Paraguay, a yield evaluation experiment on two sweet potato varieties planted in September and harvested at different times indicates that the highest yield result is obtained above 120 DPP. This suggests that, in addition to the planting period, the harvest period is also important (Soüberlich, 2017). On the other hand, Lee et al. (2016) obtained similar results, with the highest commercial tuber yield at 120 DPP, decreasing it when harvested at a shorter period. According to Cobañez et al. (2017), the planting time of sweet potatoes should be established depending on diverse factors such as the end of rainy periods or dry periods with irrigation.

Use of ridges: With ridges, sweet potato yield productivity was higher in all planting methods (spiral, vertical, and curved) and in the control group (using tubers) when compared to the experiment without ridges. Rengifo (2007) had similar results with the use of ridges, noticing a positive influence on production in three planting methods of sweet potato clones, in which he observed more commercial roots and higher sweet potato yields.

Among all planting methods with ridges, the best results were obtained with the curve, with a total yield of 18.79 ton/ha. Although the spiral method had the highest quantity of nodes, it did not obtain the best yield. It is assumed to be a result of requiring more days until harvest. However, the harvest was carried out at 120 DPP for all planting methods (Table 4 and Fig. 3).

The main benefits of using ridges are based on the increase in soil porosity, which allows water infiltration and soil aeration, and in taking advantage of more sunlight due to the elevation of the land (JICA, 2010).

Fertilizer determination: For all evaluations of different fertilizer doses (N-P-K), yield results were higher in the September 2023 planting period compared to March 2023 (Table 5). The yield response curve subject to increasing doses of potassium (K) and phosphorus (P) was similar in both planting periods. In potassium application, the yield response was directly proportional to the dose increase, obtaining the maximum value at 120 kg/ha of K in a 40-40-120 N-P-K fertilization, as shown in Fig. 4a. Regarding phosphorus, the yield increased with higher doses, giving its maximum response at 80 kg/ha of P in a 40-80-40 N-P-K fertilization However, when exceeding the indicated dose, the yield showed a progressive decrease, as shown in the response curve in Fig. 4b. Concerning nitrogen, in the first planting period (March 2023), the maximum productivity response was observed with 40 kg/ha of N in a 40-40-40 NPK fertilization, but in the second planting period (September 2023), the maximum response was observed with 120 kg/ha of N in a 120-40-40 NPK fertilization (Fig. 4c). Based on the data obtained, a formulation of 40-80-120 N-P-K in kg/ha is established, taking into account an adequate application of N for good foliar development without interference in tuber yield.

Guerrero et al. (2019), in their research to determine the effect of fertilization on sweet potato yield using N (urea), P (triple superphosphate), and K (potassium chloride) nutrients, obtained results similar to this research, but with higher yields when applying 120 kg/ha of N, 45 kg/ha of P, and 240 kg/ha of K. On the other hand, Torres and Collantes (2015) determined that the highest yield in sweet potato was attained with doses of 100 kg/ha of N, 80 kg/ha of P, and 180 kg/ha of K. These results show that the determinations and adjustments vary according to the ecoregion, but overall, there is a greater need for potassium and nitrogen than of phosphorus.

The macronutrient uptake from the soil during sweet potato harvest is serious, as indicated by Otalora et al. (2024). It results in deficient soils, with possible effects on the production and quality of tuber roots. Considering these deficiencies, nitrogen, phosphorus, and potassium supplementation should be carried out according to the needs and influence of the crop. Nitrogen supplementation in sweet potatoes influences the distribution of dry matter in the crop; however, when an excess of N is applied, the leaf area can develop more than the tuberous roots (Lebot, 2009). Regarding phosphorus, Kareem et al. (2018) point out that although the response of sweet potato to its supplementation is very low, its deficiency can interfere with growth and develop foliar necrotic lesions. Potassium plays a fundamental role in multiple essential physiological processes in the plant. This nutrient is absorbed in considerable proportions and is crucial for the adequate growth and development of tuberous roots (Cusumano and Zamudio, 2013).

적 요