Impact of forest conversion to oil palm and rubber plantations on microclimate and the role of the 2015 ENSO event

Abstract

Oil palm and rubber expansion is a main driver of the widespread deforestation of tropical rainforests taking place in South-East Asia, particularly in Indonesia. The replacement of forests with monoculture plantations of rubber and oil palm reduces biodiversity and carbon pools but also modifies canopy structure, which is an important determinant of microclimate. There is, however, a lack of quantitative information characterizing the effect of such land transformation on microclimate. We report the first medium-term observations of below-canopy microclimatic conditions (air temperature, relative humidity, vapour pressure deficit and soil temperature) across forest, jungle rubber agroforest, oil palm and rubber monoculture plantations in Sumatra/Indonesia. The data set covers a period of approximately three years (2013–2016) and includes one of the strongest El Niño-Southern Oscillations (ENSO) of the last decades. Forests were up to 2.3 and 2.2 °C cooler than oil palm and rubber monocultures respectively. The monocultures were also drier (11.9% and 12.8% less in oil palm and rubber respectively) and had higher vapour pressure deficit (632 Pa and 665 Pa higher in oil palm and rubber respectively) than the forest, while differences in soil temperature were less pronounced. Conversion from forest to other land uses, especially to monocultures, also amplified the diurnal range of all microclimatic variables studied. Jungle rubber stands out as the transformed land-use system that maintains more stable microclimatic conditions. Our results indicate that canopy openness is a key driver of below-canopy microclimate, and hence could be used in climate models to better evaluate climatic feedbacks of land-use change to rubber and oil palm. The ENSO event of 2015 led to warmer and drier conditions than in the previous two years in all four land-use systems, especially in the forest (up to 2.3 °C warmer, 8.9% drier and up to 351 Pa more during ENSO). The relative effect of ENSO was lower in the monoculture plantations, where below-canopy microclimate is generally more similar to open areas. Forests exhibited the largest differences with the pre-ENSO years, but still maintained more stable microclimatic conditions (lower temperatures and vapour pressure deficit and higher relative humidity) due to their higher climate regulation capacity. During ENSO, microclimatic conditions in jungle rubber were comparable to those in the monocultures, suggesting that while forests buffered the increase of temperature, jungle rubber might have surpassed its buffering capacity to extreme events. This capacity of buffering extreme climatic events should be considered when assessing the effects of land-use change.