Assessing the effect of landcover changes on air temperatu×re using remote sensing images—A pilot study in northern Taiwan

Extensive conversion of paddy fields to other landuse/landcover types may have adverse consequences including microclimate change. In this study we assess the effect of landcover changes on ambient air temperature using remote sensing images. NOAA AVHRR thermal images were used for surface temperature retrieval using the split window technique. SPOT multispectral images were used for landcover classification using the supervised maximum likelihood classification method. Through an inversion algorithm, landcover-specific surface temperatures were estimated. Locally calibrated relationships between surface and air temperatures with respect to different landcover types were developed using field data and used to yield average air temperatures over individual NOAA pixels. Significant linear relationships were found between average air temperatures and within-pixel coverage ratios of individual landcover types. Such relationships are inter-related and they collectively characterize the prevalent landcover conversion pattern in the region. Quantitative assessment of the effect of landcover changes on ambient air temperature was conducted by considering conditions under the prevalent landcover conversion and forced landcover conversion. Under the prevalent landcover conversion, reducing the within-pixel coverage ratio of paddy fields from the maximum of 26% to none will result in an ambient air temperature rise of 1.7–3.1 °C. Forced landcover conversion contradict the existing landcover pattern and may cause complicated consequences. Additional resources allocation and incentives may need to be introduced in order to ensure a successful forced conversion.