07/10/04 - Heavy Influence of Rainfall on Seasonal Palm Oil Production Thefactors influencing palm oil production differ fundamentally from thoseinfluencing the output of seed oils, at least as far as the seasonalfluctuations are concerned. While the seasonal fluctuations of seed oiloutput are man-made, those of palm oil are made biologically and canvirtually not be influenced by man. Oilseed crops are harvested within ashort period of 6-8 weeks and it then depends on the decision of theoilseed millers how they distribute the crushings over the whole season.The main factors influencing their decision are the demand for theproducts, the prices for the raw material and the products as well as thecrush margins resulting therefrom.As against this, the distribution of palm oil production over a year canvirtually not be influenced by the producers. It is dependent mainly onthe heavy fluctuations of rainfall in the oil palm producing areas. Onlythe annual changes in palm oil production are partly influenced by man,viz. by the changes in plantings and the resulting changes in the maturearea, though only, 2½-4 years later. Additionally any changes infertilizing and other agronomic inputs as well as the biological yieldcycle influence the annual changes in production.A look at Malaysia, the world’s largest palm oil producer, reveals howenormous the monthly changes in rainfall are. The following graphs showthat even the multiyear monthly averages fluctuated sharply between 334 mmin November and 124 mm in June. In between there is an interim peak at 235mm in April (all in West Malaysia). In assessing the effects of rainfallon palm oil production it should be taken into consideration that oilpalms are perennial crops, viz. the production process lasts throughout ayear and several years. That is, every month the following stages of theprocess occur simultaneously and continuously:- Initiation of the buds (around three years before the harvest),- Sex differentiation (around two years before the harvest), influencingthe number of fresh fruit bunches (FFB) per hectare,- Abortion of female fruits, if any (about ten months before the harvest),influencing the final FFB number to be harvested and- Flowering (about 5 months before the harvest).

All these four stages of the production process are influenced positivelyor negatively by deviations of actual rainfall from normal, but that withthe corresponding four time lags (i.e. 36, 24, 10 and 5 months) on theyields at the time of harvesting.Only if we take this complex process into consideration we can understandwhy quarterly production peaks in July/September when rainfall normally isthe lowest of the year and why it is only the second highest in October/December when rainfall normally is the highest of the year. The highrainfall in October/January means that the trees will virtually never beunder water stress during that period, so that they normally need not toabort any fruits. This is probably the major reason why the number of FFBand therefore also production of palm oil peaks ten months later, viz. inJuly/September.

The table above shows the actual rainfall in West and East Malaysia inmillimetres as well as in percent of normal. It shows that actual rainfallaveraged 88% of normal in February/August in West Malaysia and only 77% ofnormal in January/August in East Malaysia. More important than thesepercentages is the very low rainfall in millimetres, i.e. 158 mm monthlyin West Malaysia and 139 mm monthly in East Malaysia during these periods.Rainfall thus was below evaporation in six of the eight months in EastMalaysia and in four of the seven months in West Malaysia.