MODELLING AND SIMULATION OF POTENTIAL FUTURE URBANIZATION SCENARIOS AND ITS EFFECT ON THE MICROCLIMATE OF LOWER SESSION ROAD, BAGUIO CITY
Keywords: Urban Heat Islands, Air temperature, ENVI-met, Microclimate modelling, Philippines
Abstract. This paper analyzed both the horizontal and vertical extent of air temperature variations as affected by vegetation leaf area density (LAD) and building area per height threshold. The current microclimate (Scenario 1) and two potential future urbanization scenarios in Lower Session Road, Baguio City were simulated with ENVI-met. The future scenarios include the removal of the Balete trees in the median strip (Scenario 2), and removal of some pine trees and addition of new buildings in the study site (Scenario 3). Remotely-sensed data were used in creating the primary model inputs including the initial built-up layer from PlanetScope green-based built-up index (VgNIR-BI), and digital elevation models (DEM) and normalized difference vegetation index (NDVI) maps derived from unmanned aerial system (UAS) for tree canopy mapping. The horizontal variations in air temperature were observed by selecting three subsites with different proximity to the adjusted variables, while vertical analysis was done by comparing the temperature values from the near-ground up to the 20m height range. Very minimal temperature change (0 °C to −0.01 °C) was observed between the current and simulated scenarios when average temperature of the whole site is obtained. However, temperature variations were better observed with per-subsite analysis as the effects of the adjusted variables are seen to be localized and restricted only to the surrounding thermal condition. Air temperature is highest within 12:00 to 13:00 hours for the three scenarios. In the current scenario, the right region of the study site is cooler in the morning, which gradually became warmer pre-noon up to late afternoon. Removal of Balete trees with Scenario 2 have increased the temperature in the near-ground elevation only, both in Subsites 1 (up to 0.1 °C) and 3 (up to 1.6 °C), attributed to the loss of surface shadow and reduction in total LAD. In contrast, the removal of pine trees and addition of buildings impacted the air temperature across all subsites and across all atmospheric height ranges. Application of Scenario 3 to Subsite 3 which is close to the adjusted variables, gave the highest near-ground air temperature rise across the study site with a mean value of 1.4 °C and a maximum value of 1.9 °C. The ENVI-met model of the current scenario generated high accuracy (R2 = 0.59 to 0.99) which implies that the simulation outputs are reliable for this study.