Local Boundary Layer phenomena

1
Local Boundary Layer phenomena flux estimation
in LBA 2ª Reunión Nacional de Ciencias de la
Tierra UNIÓN GEOFÍSICA MEXICANA David
Fitzjarrald Jungle Research Group Atmospheric
Sciences Research Center University at Albany
SUNY, US of A
Otávio Acevedo Matt Czikowsky Jeff Freedman Ralf
Staebler Gary Wojcik Ricardo Sakai Kathy
Moore Dwayne Spiess
2
Precipitation in central Amazon is highly
seasonal (from Mendes, MS thesis UFPa, 1999)
But the rainfall at the riverside climate
stations is biased. (Garstang Fitzjarrald,
1999, p. 290) Rainfall between 4/16 5/14/87,
ABLE 2b.
Rio Solimões
3
C flux observations will always be biased by
local circulations. Which is more
important--LULC or river effects?
Confluence of the Tapajos Amazonas rivers near
Santarém.
4
It is nowhere flat, especially at night.
Topography in the LBA-Ecology region near
Santarém. Weather station flux tower sites
indicated.
5
It is nowhere homogeneous.
Surface type categories in the FLONA Tapajos
River is at left.
6
It is always awkward to do field work at remote
sites.
LBA-Ecology Pasture Site (km 77)
Tower
Solar panels
7
Humble but continuous, long-term data should be
highly prized. (But remember that information
goes both ways between modelers observers.)
Automatic weather stations near Belterra (top)
and at Fazenda Caboco, km 117 (bottom). JRG, ASRC
8
With enough data, you can select for the periods
for which any budget methods might be most
applicable. (Isolated case studies may lead
only to anecdotal information.)
On light wind days, the river breeze leads to WD
reversal.
9
Hodographs made by hour (local time) are
useful, if you can wait to get a large enough
data set.
10
(No Transcript)
11
(No Transcript)
12
(No Transcript)
13
IR GOES
Visible GOES image (recorded at USP)
14
1999 LBA-Ecology JRG USP
Average low cloud frequency for May 1999. with
the station network, we get areal distribution
of incident radiation. Looks like subtle
topography can be important in the humid tropics.
15
Freedman Fitzjarrald (2000)
16
Recovery from synoptic Disturbance--fair
weather Clouds drive the system to A preferred
LCL (or RH).
17
Freedman Fitzjarrald (2000)
18
AIRSHEDS!. Drainage outflows seen from
ripples on a calm morning. (Maryland, US of A)
Land
Water
Synthetic Aperture Radar image of Chesapeake Bay
drainage flows. (Winstead et al., 1997).
19
Lots of things go on inside the rain forest
canopy. Vertical profiles at Ducke (Fitzjarrald
et al., 1988)
20
Something completely different. Prototype of the
in-canopy sounding system, improving on the G.
Parker (right) design. T, q, CO2
21
Observing very local scale advective effects may
not be possible if there are no regular local
flows to provide a periodic signal.
Test observations done by JRG at Harvard Forest
(Staebler et al., 2000).
Which way is uphill?
22
Utility of SODAR observations. It doesnt measure
quite what we want.
HF, DRAINO (Staebler, JRG).
At Harvard Forest...
Effects of the local hill extend well into the
SBL. (SODAR results and mysteries!)
23
(No Transcript)
24
(No Transcript)
25
(No Transcript)
26
(No Transcript)
27
What we need to do Use data compositing
creatively. Deploy remote sensing
instruments to get the BL winds (e.g. profilers
and acoustic sounders) and concentrations (e.g.
DIAL) in horizontal arrays. Operate these
networks for several seasons continuously.
Develop hybrid schemes using continuous time
series from surface data. Select data for
transient periods of weak turbulence and estimate
surface flux from ?C/?t use bounding value
(envelope) analyses for guidance take advantage
of the early evening, morning transitions. Do
not ignore the role of clouds. Their presence
drives the BL toward constant relative humidity
equilibrium. Also, cloud base gives an estimate
of CBL thickness--can use ceilometer. Exploit
the nocturnal case.(!) Use the regularity of
local winds to find the advection terms. Such
natural boxes as an alpine valleys provide
predictable circulations ideal for making
composites. Drainage flows from coastal areas
carry respiratory CO2 from airsheds over water.
Measuring the volume of the outflow and its
excess CO2 allows an independent estimate of
the forest respiration rate.