Download 1 Regional Climate Model 2 Calculating Cloud Base Heights

Supplementary Material to accompany
Climate Change Scenario for Costa Rican Montane Forests
A. V. Karmalkar, R. S. Bradley, H. F. Diaz
1
Regional Climate Model
The PRECIS (Providing REgional Climates for Impact Studies) model is
derived from the third-generation Hadley Center regional climate model
(HadRM3) and is based on HadAM3P, an improved version of the atmospheric component of the latest Hadley Center coupled AOGCM (HadCM3).
It is an atmospheric and land surface model of limited area and high resolution and can be positioned over any part of the globe. Dynamical flow,
clouds and precipitation, radiative processes, the land surface and the deep
soil are all described. It is forced at the lateral boundaries by HadAM3P
GCM. The lateral boundary conditions comprise the standard atmospheric
variables of surface pressure, horizontal wind components and measures of
atmospheric temperature and humidity. The surface boundary conditions
include observed sea surface temperatures (SSTs) and sea-ice. There is
no prescribed constraint at the upper boundary of the model. The model
has 19 vertical levels, the lowest at ∼50m and the highest at 0.5 hPa with
terrain-following σ-coordinates (σ = pressure/surface pressure) used for the
bottom four levels, purely pressure coordinates for the top three levels and
a combination in between. Due to its fine resolution, the model requires a
timestep of 5 minutes to maintain numerical stability. We carried out simulations over the region of Central America (Fig.S1) at a resolution of 25
km. The model results presented in this paper are based on one member of
the ensemble.
2
Calculating Cloud Base Heights
Along the windward slopes of Costa Rica, the moisture-laden air is orographically uplifted as the trade winds encounter the cordillera. If we assume an
air parcel that cools at a dry adiabatic lapse rate without an input or loss
of moisture, then we can calculate the lifting condensation level (LCL), the
level at which a raising air parcel becomes saturated. The LCL coincides
with the cloud base level. The vapor pressure and the saturation vapor
1
pressure are calculated using the following equations.
L
e = e0 × exp
Rv
L
es = e0 × exp
Rv
1
1
−
T0 TD
1
1
−
T0 T
(1)
(2)
where, T0 = 273.15K, e0 = 0.611kP a, T is surface air temperature, and TD
is the dew point temperature. Relative humidity (RH) is given as,
RH =
e
× 100
es
(3)
Since the model gives us T and RH at the surface, TD can be calculated.
Equations (1),(2) and (3) result in the following formula for the dew point
temperature as a function of T and RH.
TD = T 1 −
T ln(RH/100)
L/Rv
−1
(4)
where, L = 2.453 × 106 JKg −1 , and Rv = 461JK −1 Kg −1 .
Now we can calculate the pressure at the LCL (mbar) using the following
formula (Geogagakos and Bras, 1984),
PLCL = 1
T −TD
223.15
+1
3.5 Psurf
(5)
where, Psurf is the surface pressure. The pressure at LCL allows us to calculate the cloud base heights using model geopotential heights. The cloud
bases are calculated for dry season months (Dec-Feb) using mean monthly
T and RH values. This exercise is to determine what the altitude of cloud
formation would be if the air from lower elevations is lifted adiabatically.
This is one of the reasons why we chose grid-points with elevation less than
800 m for this analysis. Furthermore, the horizontal grid resolution of the
model is larger in size than any individual cloud, and the model RH does
not reach 100% for individual grid-points (Fig.S2). For this reason, the LCL
estimates always lie above the grid-point elevation, which is not always true
for higher elevations. Also the altitude of cloud formation cannot be determined by looking at 100% RH surface. Therefore there is some difficulty in
estimating the actual height of the cloud bank. Nevertheless, the relative
humidity surface can be used as a grid-scale cloud cover proxy. Figure S2
2
shows the RH values as a function of elevation along a transect in Costa
Rica for the control and SRES A2 runs. These show a reduction in RH at
all elevations in future.
Reference
Georgakakos, K.P. and Bras, R.L. (1984). A hydrologically useful station
precipitation model. 1.Formulation. Water Resources Research, 20 (11),
1585-1596.
3