HAWAI`I VOLCANOES Invasion and Recovery of Vegetation after a Volcanic Eruption in Hawaii NPS Scientific Monograph No. 5

A major volcanic eruption on the island of Hawaii in December 1959 devastated an existing montane rain- and seasonal-forest covering an area of about 500 hectares (ha). The eruption resulted in a massive pahoehoe lava substrate on the crater floor of Kilauea Iki, in a new cinder cone, in an area covered with spatter and another with an extensive blanket of pumice varying along a fallout gradient from over 46 m to less than 2 cm deep. Six new habitats were recognized by kinds of substrate and remains of former vegetation. A study was made of plant invasion and recovery from the time of the disturbance till 9 years thereafter. Plant records consisted primarily of periodically listing species by cover-abundance in a large number of quadrats along a transect system that crossed the crater floor and extended about 3 km along the fallout gradient. The atmospheric environment was studied concurrently by records of rainfall, lateral rain- and steam-interception, and desiccating power. The substrates were examined for their soil moisture properties, temperatures, mineralogical properties, and available plant nutrients.

It was found that plants moved onto the crater floor within the first year. They progressed concentrically towards the crater center in correlation with a substrate-heat gradient that cooled progressively from the margin inward. Plant invasion on the cinder cone was delayed by 2-3 years, because of prolonged volcanic heating from below. A fast invasion took place on the spatter habitat where a surviving rain forest was nearby and where tree snags provided additional moisture locally at their base by intercepting wind-driven rains. Establishment at snag bases was also noted on the pumice, and, generally, plant invasion occurred by aggregation of plants in favorable microhabitats which included crevices and tree molds. On the pumice, invasion progressed at a relatively uniform rate in spite of differences in substrate depth and atmospheric environment. The increase in plant cover was much faster on the habitats with vegetation remains than on those without. On the latter, the plant cover was still insignificant in year 9 after the eruption, in spite of a near total spread of plants across these habitats.

The sequence of life form establishment on substrates without vegetation remnants was clearly algae first, then mosses and ferns, then lichens, then native woody seed plants, and finally exotic woody and herbaceous seed plants. On the substrates with former vegetation remains, exotic seed plants participated in the invasion process from the beginning. This was related to the availability of microhabitats with water relations favorable for plants with normal root systems and probably higher water requirements than the native selerophyllous woody plants. A remarkable recovery occurred among Metrosideros polymorpha trees that were buried up to and over 2.5 m deep under pumice. Several native shrubs resprouted after their entire shoot system had been buried. The best herbaceous survivors were those with underground storage organs, which included both native and exotic species.

The invading exotics in no way interfered with the establishment of the native pioneer plants. Initial stages of succession were observed whereby native woody plants began to replace exotic woody plants. Among herbaceous plants, exotic species were far more numerous, because there are only very few native species in this group. A succession, in part caused by competitive replacement, was noted among the exotic herbaceous plants. Thus, there appears to be no threat of native plants to be replaced by exotics on these new volcanic substrates. The native forms are better adapted to these harsh environments. But exotic complementary life forms are expected to remain in association with the native vegetation because of a lack of life forms among the native species to fill the available niches.