Southwestern Free State product trial under irrigation at Petrusburg, 2024 to 2026

Estimated reading time: 9 minutes

The Southwestern Free State potato production region produces around 1.86% of the country’s commercial potatoes on 954 ha. The most prominent cultivars produced as part of this region’s main harvest for table and processing purposes are Sifra (89%), followed by Mondial, Innovator, and Panamera.

Petrusburg is located in South Africa’s dry continental region (Figure 1). The farm where the trial was conducted has recorded an average annual rainfall of 565 mm over the past 25 years, as per the local Agricultural Research Council (ARC) weather station. This region is characterised by very hot summers and cold winters, with frost occurring from June to August.

Trial design and details

The product trial at Petrusburg is laid out in a random block design with three replications per cultivar. Table 1 contains extra information relating to the trial. Soil samples were taken prior to planting in order to determine the soil nutrient profile of the trial site (Table 2).

The product trial includes a variety of companies with diverse programmes. As dictated by policy, this article refers to nine programmes (treatments) tested in the trial over three years. In all three trial seasons the relevant programmes were tested in addition to the farm’s standard programme (control) using the cultivar Sifra.

As in the case of the Petrusburg product trial, the evaluation of treatments provides results relating to yield and marketing indices, among others. The goal of the trial is to determine how the programmes in question performed within the region.

It is very important to note that the programmes cannot necessarily be measured against each other, but rather against the control. Various programmes deliver different outcomes, and the producer can discuss the finer details of each programme with the relevant representative.

The standard spraying and fertilisation programmes employed on the farm served as the control treatment in the trial. Therefore, this article includes the control treatments as well as the programmes specified in Table 3.

Sorting and classing

The marketing index of the treatments is calculated by classing and sorting each treatment according to quality and size groups (such as Class 1 large or Class 2 large-medium). All three replications from this trial were combined, washed, and then classed and sorted by the packhouse.

Prices were then compared to market prices at the time of harvesting. The performance of the treatments cannot be based on the results of only one specific season, since climate and tuber quality can differ from one year to the next, which is why programmes are preferably tested over at least three seasons.

Rainfall and temperature

Factors such as temperature, water availability (whether through good irrigation scheduling or rainfall), and heat units have a significant influence on the potato plant during its growth period. These factors are considered when evaluating treatment performance.

Applicable daily weather data for the relevant season is obtained from a weather station nearby. The long-term data consists of records from the nearest ARC weather station, located 9 km from the trial.

Figure 2 illustrates the rainfall trend of the past three seasons. High rainfall was recorded in the first months after planting in seasons one and two. Seasons two and three recorded excessive, above-average rainfall during tuber fill in the latter half of the season, which may have led to quality issues and an increase in Class 3 and rejected potatoes.

Minimum and maximum temperatures are illustrated in Figures 3 and 4. The collection of heat units during a growing period is an important factor in the development of a potato plant. The trend of heat units available for the past three seasons appears to be significantly higher in season three compared to the long-term average of heat units accumulated in the first half of the growing season. All three seasons recorded substantially more heat units than the long-term average.

Yield across three seasons

Yield data recorded during harvesting is subjected to statistical processing using the GenStat® program.

The Tukey test of least significant differences (LSD) was used to separate the mean. The treatment effect during this trial (Figure 5) was statistically significant (p<0.05) and the coefficient of variation (CV) was consistently low (8.3, 7.2 and 11%, respectively). These factors indicate that the trial was well executed and the results are therefore reliable.

The yield of each of the cultivars is divided by the trial average, with the average of all the cultivars taken as 100%. This is used to determine a yield index, and each cultivar’s performance in terms of yield is taken as a percentage of the trial average.

The average yield of the trial for the three seasons was 93.74, 109.6 and 66.56 t/ha, respectively. The above-average performance of the first two seasons can be attributed to three factors. Firstly, the trial could not be conducted on soil with no history of potato production. Secondly, the trials were planted at a slightly higher plant density/ha than previously.

Thirdly – and probably the greatest contributing factor – was irrigation using water of excellent quality with a significantly lower sodium content.

Unfortunately, season three was characterised by poor emergence across the trial site, which gave rise to yield losses.

Few statistically significant differences were observed in seasons one and two. In season three, which recorded lower yields, however, there were clear differences, with Bayer 1 and 2, FMC, Bancella, and GuanoBoost producing the highest yields (Figures 6 to 8). It is important to note that, given the varying and different outcomes of the programme, the treatments cannot necessarily be compared with one another.

The size distribution of the various treatments is compared in Figures 9 to 11. Fewer large potatoes were produced in seasons one and three than in two, which generally delivered a very high yield. Despite the high yield, however, season two produced proportionally fewer Class 1 potatoes.

Given the nature of the seasons, the performance of treatments will vary from one season to the next, simply because climate varies from one season to the next. Thus, it is important to consider consistent treatment performance across a number of seasons, rather than relying on the good performance of a single season. Bayer 1 and 2, Karabos 1, and FMC showed the least variation for the past three seasons in the Petrusburg trial (Figure 15). – Enrike Verster, Potatoes SA, and Johan Odendal, producer

Special thanks to Lushof Boerdery and each participant in this project, as well as the staff involved. For more information, send an email to enrike@potatoes.co.za.