Long-term use of potassium fertilizer for vegetable crops in the upcountry intermediate zone

A long-term field experimentwas conducted a t Bandarawela on a Red Yellow Podzolic soil during the ten cropping seasons commencing Maha 86/87. Four levels of potassium a t rates of 0,25,50 and 100 k g q O / h a were appl~ed for each crop. Potato, cabbage, tomato, pole bean, cabbage, potato, cabbage, tomato, pole bean and cabbage were cultivated in this sequence: ~ i r s t three crops potato in Maha 86/87, cabbage in Yala 1987 and tomato in Maha 87/88 did not show any response, even a t 100 kg &O/ha application. This may be due to the high inltial K content of the soil. Thereafter there was response up to 25 kg qO/ha application for the pole bean in Yala 1988 and cabbage in Yala 1989 crops. Potato in Maha 89/90 gave significantly highest yield a t the level of 50 kg K,O/ha. However, cabbage in Yala 1990, tomato in Maha 90191, pole bean in Yala 1991 and cabbage in Maha 91/92 crops showed response to 25 kg qO/ha and even up . to 100 kg &O/ha. Crops such as potato and pole beans can be categorized as "low K-removal" and cabbage as "high K-removal".After cultivationof low K removal crops such as potato and pole bean, the soil showed an increase of exchangeable K when 100 kg &O/ha was applied. Increase of soil K was seen when potassium is continuously applied to the same soil a t the ratf: of 100 kgK,O/ha but the make up of the cropping sequence is a n important factor to be considered. I t appears that high yields of vegetables such as potato, tomato, cabbage and pole bean, could be obtained with 100 kg &€)/ha per season application without any buildup of K in the soil. These results suggest that the application of 100 kg %O/ha to vegetable crops such as potato, tomato, cabbage and pole bean per season is is necessary and it avoids the build up of high soil K i n this soil with the high yields.


INTRODUCTION
The Upcountry Intermediate Zone of Sri Larika lies between 600 and 1400m above mean sea level. The topography comprises steeply dissected hills with rolling terrain. The mean minimum and maximum temperatures are about 15 and 270C respectively. Average annual rainfall ranges between 1100 to 1400 mm. The soils are mainly'Red Yellow Podzolicl'with a pH range of 4.0-5.5 The hill slopes are planted with potato and vegetables in rotation throughout the year. In the lowlands, rice is planted during January-March and harvested during May-July. Potato and vegetables follow rice in these lands upto the end of December. Almost all farmers in this region use inorganic fertilizer mixtures "'present address: Department of Agriculture, Peradeniya. in combination with organic fertilizers such as cattle and poultry manure. The agriculture in the upcountry areas of Sri Lanka is considered as intensive cultivation of potato and vegetable crops. Use ofhigh analysis fertilizers at high levels in this region has contributed towards the accumulation of plant nutrients in the soil.2 Potassium is one of the essential elements in the plant food, and one of the three that is generally in short supply in the soil, and limiting crop yields. Hence it often needs to be added regularly as fertilizer. It has been recognized that potassium is needed by the plant in large quantities for generating high yields with improved quality of produce. Potassium is important in photosynthesis, the basic process in plant production by which the energy from sunlight is converted to carbohydrates. The results reported by Skogley3 have shown the serious consequences of K deficiency on plant development in areas where warm and dry winds are common during the growing season. In addition, potassium speeds up the flow of assimilates, favors protein formation, improves the effect of N, increase water-use efficiency and tolerance to salinity.' The use of results from experiments of short duration has sometimes led to mistaken fertilizer advice. As farming is a long-term activity, experimentation should be similarly long term.5 When the soil K content is already high, there is little or no effect of K fertilizer on crop yields in the first few seasons of an experiment.= Most tropical soils are highly weathered, and many are acidic and low in bases. They have a low cation exchange capacity and are unable to adsorb large quantities of K from fertilizers on the exchange complex.' The total K requirement of the crop must be considered when determining the fertilizer requirement of a crop. The improved varieties of plants with high yield potentials generally take up more K than traditional varieties and require a correct balance between N, P and K fertilization for realizing optimal yield^.^*^ Large quantities of potassium are utilized by the intensive systems of vegetable production, and to ensure a high yield from such crop sequences, soil fertility has to be maintained at high levels.
In year round commercial agriculture which prevails in the upcountry of Sri Lanka where the crops are high yielding, soils may rapidly be depleted of K, unless its removal is adjusted by the addition of K fertilizers on a regular basis. On the other hand, excessive additions of potash may lead to a K build up in soil which may even lower the magnesium availability to There is hardly any reported experimental evidence on the effect of longterm application of K on the yield of vegetable crops in the Upcouniiy Intermediate Zone of Sri Lanka. A field experiment was therefore conducted to investigate the long-term effect of fertilizer K on a cropping sequence that is being followed in this region.

METHODS AND MATERIALS
A long-term field experiment was conducted on a Red Yellow Podzolic soil (Table 1) a t the Regional Agricultural Research Centre, Bandarawela, where four levels of Potassium (0,25,50 and 100 kg &O/ha) were tested on a sequence of crops in a randomized complete block design with six replicates. The plot size was 3 m x 2.5 m. Nitrogen and phosphorus levels were kept constant a t 200 kg N k a and 150 kg P 0 /ha respectively, for cabbage, tomato and pole bean. The  Spacing, dose and time of application of fertilizer for each crop is shown in Table 2. Nitrogen was applied as urea, P as concentrated super phosphate and K as muriate of potash.
The crops were grown under rainfed condition with supplementary irrigation whenever necessary. The fields were maintained in a weed free condition throughout the experiment.
Six composite soil samples were collected from each replicate before commencing the experiment. The chemical analysis of soil from the experimental field at the commencement of the experiment is given in Table 1. Soil samples were also collected from each plot after each crop and analyzed for NH,OAc extractable K.

RESULTS
The effect of addition of potassium on the yield of vegetable crops is given in Table 3. Potato in Maha 86/87, cabbage in Yala 1987 and tomato in Maha 87/88 crops did not show any significant yield response to addition of potassium. However, application of potassium at 25 kg &O/ha resulted in a significant yield increase for pole bean in Yala 1988 and cabbage in Yala 1989 over the control. Application of 50 and 100 kg &O/ha did not show significant increase in yield over 25 kg qO/ha during the above two seasons.
Potato in Maha 89/90 gave a significant yield response to addition of 50 kg &O/ha, but cabbage inYala 1990, tomato in Maha 90191, pole beaninyala 1991 and cabbage in Maha 91/92 showed significant yield response to the addition of 100 kg &O/ha.

DISCUSSION --.
If the experiment was conducted for only three seasons an erroneous conclusion may have been drawn that fertilizer Kis not required for crops grown on this soil. This study demonstrates the importance of long term experimentation in the fertilizer response studies. It may be suggksted that 100 kg &O/ha per season may be a suitable level of fertilizer application for year round Table 3: Effect of potassium on the yield of crops grown in a sequence (ma).   . .

'
Based on data given in Table 5. The lack of response to potassium during the first three seasons and the positive responses obtained later may be understood by considering the exchangeable soil K content determined a t the end of each season (Fig. 1). The soil K content of the control plots as well as those receiving 25 and 50 kg &O/ha continued to decline after the third season except a t the end of the ninth season for the 50 kg &O/ha treatment. The soil potassium content of the control treatment dropped upto 0.1 md100 g a t the end of the sixth season. The yield responses obtained to addition of K fertilizer during the fourth season may be attributed to a lowering of soil K values in the control plots. The study has clearly demonstrated the effect of crop removed on the depletion of soil K level. The amount of K removed by potato, bean, cabbage and tomato are shown in Table 5. Using these values the Kremoval by crops during each of the ten seasons is given in Table 4. Although bean crop removes more K than cabbage on a per ton basis ( Likewise, soil K content did not show much decline or even increased occasionally with addition of 100 kg &O/ha during seasons when comparatively low amounts of K may have been removed from soil as with pole bean in seasons 4 and 9 and potato in season 6. These results show how the potassium content in soil can be rapidly influenced by the type of cultivated crop. Hence, potassium fertilizer should be applied in accordance with the needs of the different crops.12 Taking into consideration, the results of this experiment, it could generally be considered that in a cropping sequence, a crop that removes low quantities of K by virtue of its yield potential, when followed by another crop in the sequence will normally give high yields with low quantities of applied K due to the accumulation of exchangeable K from the previous crop. On the other hand, where crops remove high quantities of K ,application of potash is necessary for succeeding crops in order to maintain the K status of the soil. The results of the experiment reported here suggest that application of a t least 100 kg &O/ha per crop is needed for vegetable crops grown in this region. This will be an economical approach to the farmers who generally apply almost 3-4 times as much as this amount. Therefore, it is obvious that even a t fluctuating vegetable prices and ever increasing fertilizer prices in Sri Lanka, there is economical benefit to the farmers if they apply 100 kg qO/ha.