Developing Land Use Change Climate Change Scenarios

1
Developing Land Use Change Climate Change
Scenarios
For Central America
Presented by Bob Oglesby, NASA/MSFC
2
Project Hypothesis Deforestation and other lan
d use changes currently occurring
in Central America have a profound, and in many
cases detrimental effect on the hydrologic cycle
, increasing the probabilities of both devastati
ng droughts and floods.
The goal of this research is to understand the
effects of human activity upon the hydrological
processes over Central America in order to
understand and so ultimately forecast climatic
effects for decision making by scientists, educa
tors, and policy-makers.
3
KEY QUESTIONS How significant to Central Ame
rican climate change is deforestation / land use
change in comparison to natural climate
variability? What role do these anthropogeni
c factors play in floods and droughts?
What are possible future scenarios of climate
change resulting from greenhouse gas
increases and how do they compare to changes
projected to result from deforestation
and other land use changes?
4
Deforestation is not the only agent of climate
change over Central America. Globally, warming of
surface temperatures, with associated changes in
the hydrologic cycle and other important aspects
of climate, is thought to be occurring due to the
steady rise in greenhouse gases (GHGs).
Indeed, deforestation has a direct impact on t
his by causing an overall net release of
GHG to the atmosphere But how will these fut
ure climate changes affect Central America? Will
deforestation make them worse? Or conversely,
will the climate changes help ameliorate the
impacts of deforestation?
5
Though these effects are very difficult to
quantify, and hence predict, it is my
personal belief that the recent floods and
landslides in southern Mexico, El
Salvador, and especially Guatemala were
exacerbated, if not directly caused,
by deforestation of very steep hillsides.
The rain that fell has been described as a ste
ady, moderate rain, people apparently were caught
by surprise by the extent of the flooding and
land- slides. If the hillsides had been fores
ted, much more of this rain could have been
caught and absorbed. It would have been released
more slowly into streams, and the hillsides them
selves more likely would have remained stable.
Again, this is my personal conviction which I ca
nnot at this time prove.
6

• We have already performed some preliminary
  studies relevant to our proposed research.
  
• This prior work has provided two key insights
• We have shown that a regional climate model (the
  NCAR/PSU MM5) does a credible job at simulating
  at least precipitation over Central America.
  
• We have shown that complete deforestation of
  Central America has the potential to have major
  impacts on temperature and precipitation. This
  experiment was meant to be an end-member test
  to bound the possible climate impacts.
• We suspect other facets of the hydrologic cycle,
  e.g., evaporation and runoff, will also
  eventually be shown to be similarly impacted.
  

7
(No Transcript)
8
(No Transcript)
9
The next slide shows the impact of human activit
ies on climate, especially as they impact
drought. We made two runs using the MM5 region
al climate model. One run had all of Mesoamerica
containing evergreen forest, while the other run
had all of Mesoamerica containing grassland.
Replacing trees with grassland has two major eff
ects 1. an increase in albedo, leading to cooli
ng and stabilization of the atmosphere
2. a large reduction in evapotranspiration from
the surface, leading to warming and stabilization
of the atmosphere. The reduction in
evapotranspiration overwhelms the increase in
albedo, leading to warmer surface conditions.
But both processes tend to stabilize the atmosph
ere and reduce precipitation
10
Mesoamerican Deforestation Scenarios
Implications for Wet and
Dry Seasons
January (dry season) Temperature (left) warms
everywhere but only by 1-2deg C Rainfall (right)
- little
impact as its the dry season!
July (wet season) Temperature (left) warms every
where, but now as much as 2-5 deg C. Rainfall (
right) shows a decrease of 20-30 over much of t
he Mayan region
11
We see warming and drying in both the dry and wet
seasons, but the impact is much greater in the w
et season for both temperature and rainfall.
These results suggest that human activities
in this region could help trigger,
prolong, or exacerbate drought.
12
The next slide shows the possibility of natural
drought. We show time series of warm season ra
infall (defined as rain from June 1 through
September 30) as averaged over the Mayan lowland
region of Guatemala and the Yucatan. The rainfall
was simulated by a global climate model, the
National Center for Atmospheric Research CCSM3.

13
This plot shows rainfall over the Maya
lowland region for the wet season
(June-September) for 100 years of
present-day conditions as simulated
by a global climate model.
This plot shows rainfall for the wet
season for a model simulation of the
period 1870-1999. Note much more prolonged period
s of drought, e.g., 1952-1981, during which 24 o
f 30 years have below average rain.
14
The plot on the left is from a 100 year
present-day control run, that is with
present-day conditions run over and over again
for 100 model years. We see many dry periods,
though only a few extend for more than a few
years. The plot on the right is from a run for
the period 1870-1999, that is, all appropriate
forcings (e.g., changes in solar irradiances,
greenhouse gas increases, aerosols, volcanoes,
etc.) for this period. They key difference is
that now we see some prolonged periods of dry
conditions, including one period of 24 out of 30
years being dry These results suggest that natu
rally-occurring drought can wreak havoc on this
region. The key questionis whether human-induced
land use changes can make these droughts worse,
or occur more frequently!!
15
The next step is to modify the current land
surface classification according to the estimates
of present-day (circa 2002) deforestation and
other land use changes. We will also make runs
projecting rates of deforestation forward for up
to 20 years into the future. That is, we will
make model runs with more realistic scenarios of
deforestation (and other land use changes).
The land use classification portion of the SERVI
R project will provide us with the present-day
landuse classifications. In conjunction with ou
r Central American partners, we need to come up
with likely future scenarios of land use
changes. (To help accomplish these, I have been
spending time in the land use workshop being
held Concurrently with ours.)