Soil fertility in the tropics can be influenced by landscape and precipitation

A new study conducted in the Hawaiian Islands has revealed that landscape and erosion play crucial roles in determining soil fertility in tropical ecosystems.

“This study is the first to accurately predict the distribution of nutrients across a complex tropical forest landscape, and then to detect these shifts in nutrient status using airborne sensors,” says Stanford University graduate student Stephen Porder, lead author of the study, which will be published in this week’s online edition of the Proceedings of the National Academy of Sciences (PNAS).

Porder’s co-authors are Carnegie Institution scientist Gregory P. Asner, an assistant professor (by courtesy) of geological and environmental sciences at Stanford; and Peter M. Vitousek, the Clifford G. Morrison Professor in Population and Resource Studies at Stanford. “Tropical soils often are assumed to be highly weathered and thus nutrient depleted,” the authors write. But the study, which focused on the island of Kauai, revealed a complex “biogeochemical patchwork with almost equal areas of high and low nutrient availability.”

Phosphorous, calcium and other minerals essential for plant growth are derived from the breakdown of bedrock as it is converted to soil. “As soil ages, this bedrock source is thought to be depleted, and these elements become increasingly scarce,” Porder explains. “However, this process has been studied almost exclusively on uneroded surfaces, and thus it is uncertain how well it applies to actively-eroding landscapes.”

In previous studies, Porder, Vitousek and their colleagues discovered that the erosion of mineral-rich rocks actually rejuvenated soils along several slopes in Hawaii, producing higher phosphorous concentrations.

The PNAS study was designed to determine the extent of nutrient rejuvenation in old Hawaiian forests. To find out, the researchers focused on a 7.7-square-mile area of Kauai that includes stable ridgetops, eroding slopes and valleys that receive very different amounts of annual rainfall.

The researchers first collected soil samples and leaves from native ohia trees along a series of transects running from hilltops to valley bottoms. The leaves were analyzed for their nutrient content, and that information, along with rainfall and slope data, were used to develop a map predicting phosphorous concentrations throughout the study area. The researchers predicted that stable ridgetop soils would have little phosphorous, while low-lying slopes and valleys were expected to contain large concentrations of the nutrient.

Those predictions were then tested with a remote sensing technique developed by Asner and Vitousek to analyze the chemical content of leaves in a forest canopy. The researchers flew over the study area in an aircraft specially equipped with an infrared spectrometer that detected the chemical concentrations in ohia leaves. The results of the flyover confirmed what the scientists had predicted–namely, that only about 17 percent of the landscape is nutrient-poor. It turned out that, while upland regions tended to be depleted of phosphorous, the slopes and valley bottoms were often as fertile as the most verdant forests in the Hawaiian Islands.

“Our study makes two important advances in the understanding of nutrient availability in tropical soils,” Porder says. “First, it builds an explicit prediction of where in the landscape nutrient availability is likely to be high and low. Second, it uses remote airborne sensors to detect phosphorus concentrations in the forest canopy and independently confirms our predictions.”

The authors point out that these results may apply to other tropical ecosystems, noting that about 10 percent of all tropical ecosystems “have slopes greater than 10 degrees, with higher percentages in Central America, Asia, parts of South America and some Pacific islands.”

They note that, while topography and soil age are the dominant drivers of soil fertility in their study area, other factors–such as tectonic uplift, vegetation and climate–are likely to produce a similar patchwork of nutrient availability in the tropics.

“Our results indicate that not all old, tropical forests are nutrient-poor,” Porder says. “Understanding natural variability in forest soil fertility is critical for understanding the response of tropical forests to climate and land use change. Hopefully our study will spur additional and much-needed research into the spatial patterns of nutrient availability of other tropical landscapes.”