Pedestrians' thermal sensation is affected by exchange of energy with their surroundings, mainly through radiation and convection. Ground-cover vegetation can have an important influence on this energy exchange. However, not all plants are the same, and although most plants have a similar albedo, water-efficient plants such as succulents may have substantially higher daytime temperature than grass or broad leafed species. Modelling pedestrian comfort thus requires a means of characterizing different plants and their effect not only on air temperature and moisture, but most importantly on radiant exchange.
The study proposes a new method for estimating the surface temperature of surface cover plants such as grass, creepers and succulents that is based on an adaptation of the sol-air temperature. The sol-air temperature' offers a close approximation of the actual surface temperature and is defined as the equivalent outdoor temperature which will cause the same rate of heat flow at the surface and the same temperature distribution throughout the material as results from the outdoor air temperature and the net radiation exchange between the surface and its environment.
The green sol-air temperature' introduced here adds the effect of evapotranspiration:
where Ta is air temperature, ΔR and ΔM are net radiant exchange and net mass transfer (evaporation or condensation) respectively, and hc and hv are surface transfer coefficients for heat and water vapor. The third term on the RHS of the equation is added to the first and second terms that were part of the previous formulation. ΔR is the sum of the net shortwave component αK↓ and the net long wave radiation εsL*, where K↓ is incident solar radiation, α is absorptivity, εs is the emissivity of the surface and L* is the net long wave radiation. ΔM is given by ρλ(q*)/(RaV-Rs) where ρ is air density, λ is the latent heat of vaporization of water, q* is the vapor pressure deficit and RaV and Rs are the aerodynamic and stomatal resistances of the plant canopy. hc and hv are functions of air speed at the surface, and are assumed to be approximately equal. The model in this form is applicable to ground vegetation that covers the entire area of soil.
The green sol-air' is derived from a simple, single layer model with one effective surface, similar to so-called giant leaf' models used in large-scale climate models. It is suitable for dry foliage, and ignores both physical and chemical heat storage, as well as heat and mass transfer processes within the canopy.
The utility of the model was tested by comparing the predicted green sol-air temperature for different ground cover plant types, as well as dry bare soil and asphalt, with surface temperatures of the respective plants obtained (separately) by infra-red thermography. The ground cover plants, which were chosen according to their compatibility to arid conditions and their potential for rapid growth and year-round appearance, displayed a wide range of temperatures in sunny conditions, providing a good test of the model as well as indication of its importance.
For example: On a typical clear day with global solar radiation reaching a maximum of 970 Wm-2, the average surface temperatures of all vegetated plots were higher than air dry bulb temperature, but considerably lower than those of the non-vegetated surfaces. The temperatures of the three CAM-type succulent plants (malephora, aptenia and drosanthemum) were higher than the leafy plants (kikuyu grass, lippia and convolvulus) by approximately 10°C at midday, with all three succulents reaching temperatures above 45°C, compared with an air temperature of 31°C. Bare ground temperature was some 15°C warmer still, reaching almost 60°C.
Using meteorological parameters obtained from a weather station about 1 km. away, and introducing appropriate values of Rs, the error in estimated surface temperature during the daytime hours over three measuring periods for all plant types was less than 2°C.
The effect on pedestrian thermal comfort of ground cover plants is manifested to a great extent through modification of the mean radiant temperature (Tmrt). This may be illustrated in the case of a pedestrian standing in a hypothetical open space at 30N latitude with a sky view factor of 1' and a ground surface an albedo of 0.25. On June 22, assuming a solar flux of 750 Wm-2 and an incoming IR flux of 315 Wm-2, the difference in Tmrt impacting on a pedestrian due to an increase in surface temperature of 10 C (grass at 35oC to succulents at 45oC) would be over 4 degrees (from 44.4 to 49.1oC). For a wind speed of 1 ms-1 and RH=50%, this translates into a change of the UTCI from 34.0 to 35.2 degrees.
A change of this order in the UTCI value does not change the classification, which in this case remains strong heat stress'. However, it adds to the uncertainty in estimating other components of the radiant balance, and users of the Index should be provided with a means of incorporating the effect of plants in their calculations. In the conditions illustrated, a paved surface with a similar albedo (0.25) would have a surface temperature of about 60oC, giving Tmrt=61oC and UTCI=38.2. A difference of over 4 degrees compared to a similar grassy surface is too large to ignore.
- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner