Introduction

The most common rat species associated with young oil palm plantations in Costa Rica is Sigmodon hispidus , which normally feeds on leaf bases and the bulbs of the palms. However, when the rat population is very large, the rodents climb up to the young bunches where they feed directly on the developing fruits. Zigodontomys sp. is another rat species found in oil palm plantations, but its importance has not yet been documented.

In general, the life cycle of most rat species is between 10 and 22 months, and these are extremely prolific animals: sexual maturity is reached in about four months, and a female may produce a litter of 5-10 youngs every 3-4 months.

We do not fully understand the reasons for the sudden increase in rat populations, to levels where they may cause serious damage to agriculture. Such population explosions and the resulting damage normally occur every five years, and must be counteracted by costly pest-control measures to protect the young palms and the first bunches.

The damage caused by H. hispidus in oil palm is more severe during the first 3-4 years after planting. There are at least three factors that contribute to this:

• H. hispidus is not a very good climber, so damage to the bunches decreases as the stem grows in height, 
• New oil palm varieties are very precocious, so first bunches are produced very close to the ground. 
• The cover crop (kudzu) and other weeds provide many hiding places for rodents.

During 1999, a dramatic increase in the rat population was observed in young 2-3 year-old oil palm plantings over an area of approximately 750 hectares, in the Quepos region of Costa Rica (Central Pacific). Most of the damage was inflicted on the young developing bunches, causing a 3-6 month delay in the start of harvesting.

The over-population of rats was evident in the change in sex ratio from the usual 1:1 to five females for every male. In addition, many animals were active during the day, contrasting with the nocturnal habits of H. hispidus .

Index of captures. Mechanical rat traps, spaced 10 meters apart, were placed along transects, using a minimum of 100 traps/ha. The success of this campaign was defined as the ratio (percentage) between the number of captures and total number of traps used. This ratio was above 60% ( Fig. 1 ) during the period of most severe damage. For many crops such as rice and sugar cane, a value of 10% in trap efficiency is already considered high, but the critical value for oil palm is possibly much higher (around 30%). A percentage of between 60 and 80% of activated traps was associated with 30-40% damage to the young developing bunches (first year of harvesting).

Management . It was essential to adopt an integrated approach that included habitat management and biological and chemical control components. Habitat management included the search for and elimination of hiding and breeding sites, such as piles of organic matter and other materials, and poorly-maintained drainage slopes. Weed control was improved both around the palms and between them. Another factor that was considered was the migration of rodents from adjacent fields where weed control was deficient. 

In the replanting areas, the piles of trunks from the previous plantation provided particularly favorable sites for hiding and breeding, making the attacks on young plants even more severe. In these areas it was necessary to carry out a phytosanitary pruning earlier (first year harvesting), which involved cutting off old leaves and the bunches closer to the ground.

Rats have many natural enemies, but commercial oil palm plantations normally do not provide an appropriate environment for the activity of this beneficial fauna. Moreover, their impact is even less when other environmental factors promote a disproportionate increase in rat population. However, some strategies encouraged predator activity, particularly raptors, such as placing observation perches, in the form of high wood piles over the palm canopy. Keeping the palm circles clean also facilitated raptor activity. Nesting facilities for the barn owl Tito alba have not been yet used in Central America.

An inexpensive anticoagulant bait was made from crushed corn (46 kg), Racumin powder (3 kg), and sugar cane molasses (1 liter). The components were mixed (using gloves and a suitable mask), and put into small plastic bags (15-25 grams/bag). Before placing the bags in the field, they were sprayed with coconut oil to mask human odors left during handling. These baits were placed near the breeding sites and along the trails used by the rats (they should not be placed in open areas since rats avoid such areas to remain out of reach of predators). 

When labor was scarce, the Racumin baits were substituted with brodifacoum baits, which are offered commercially as ready-to-use products. In badly affected areas up to one bait per palm was used, and in not so severely affected sites, baits were placed in the first 3-4 rows along the border of the lots.

In general terms, the cost of controlling the rodent damage described above was around US$25/ha, and the rat population was brought down to economically-acceptable levels within about eight months.