Why not simply apply fertilizer, compost or mulch to restore fertility to eroded tropical soils? We found problems with these approaches.

Does fertilizer improve the soil for tree seedlings?

Chemical fertilizer

The apparent solution to poor soils is to fertilize them, so we did several experiments using chemical fertilizers in the field and in the tree nursery.  In addition, five of the 45 plots in Bloques '93  tested the effect of fertilizer on the native timber tree Terminalia.


The fertilizer recommended by local foresters is 10-30-10, a formula high in the three nutrients--nitrogen, phosphorous, potassium, respectively--that plants need the most.  All three tend to be scarce in tropical soils before erosion and are severely limited in exposed sub-soils after erosion. 

We put 50 grams of 10-30-10  in the planting hole the first year. The second and third years we also followed foresters' recommendations and applied a more complete fertilizer ("formula completa") in three holes dug around the root zone of each seedling.

Sometimes fertilizer seemed to help early seedlings grow better, but their advantage compared to unfertilized controls was always lost by the sapling stage.  We think fertilizer stimulates the growth of weedy competitors better than the trees themselves.  Much fertilizer likely runs off in heavy rains.  The cost therefore is not worth it to the farmer.

Mulch fertilization

Although chemical fertilizer seems to have no lasting positive effect on tree seedlings, mulch compost might, especially mulch prepared from leaves with high nitrogen content, such as legumes, or mulch that decomposes relatively slowly releasing nutrients over time.

In the wet Tropics, we don’t need to compost plant matter because it changes chemically in hours due to high heat and humidity.  We tested the effect of various tree and herb mulches, using grass mulch as controls, on five species of trees.  We cut leaves from each source and standardized the amount of dry

Maria seedling with Vochysia mulch a few days after application.  Note netting that holds the mulch in place.

Maria seedling with Vochysia mulch a few days after application.  Note netting that holds the mulch in place.

material applied to the seedlings. See the first two photos in the panel below, preparing mulch (left) and weighing fresh weight (second photo).  Heavy rainfall on steep slopes carried away the mulches, so we learned that we had to stake down the mulches around each seedling with netting.  Analysis suggested that one species of mulch (Vochysia) helped several of the target species.

  • Mulch decomposition: Since the effect of mulch on nutrient status of seedlings should depend partly on rate of decomposition, we measured these rates in the species used in this experiment--see third photo from the left, above.  Rates were so variable in the field that the different species of leaves did not differ statistically.
  • Vochysia leaf fall:  If Vochysia mulch does help other species, then the ability of the natural mulch from decomposing leaves to help nearby trees will depend on the rate of leaf fall from Vochysia in the field.  Therefore, we measured leaf fall by capturing leaves in nets under trees (fourth photo above).

Mulching was labor intensive since it had be cut from healthy plants, chopped, and carried to the site.  We prevented it from moving downslope away from the seedlings in heavy rains by staking down the mulch with netting, a time-consuming work that most people wouldn't be able to do on a large scale .  Therefore, we concluded that the benefit of applying mulch does not outweigh the cost in labor, especially in steep terrain. 

However, mixing tree seedlings with living plants whose leaf fall provides mulch might offer an alternative, which we .test in our experiments, 'Facilitation by Plants'.