Linking water and climate change: a case for Brazil Eunjee Lee Sustainability Science fellow, Harvard Kennedy School with Prof. Paul Moorcroft, Angela Livino and Prof. John Briscoe
Outline 1. Overview: The Water Cycle 2. Linking water between land and atmosphere Impacts of reduced precipitation on rainfall-reliant sectors 3. A case for Brazilian hydropower: Energy sector 4. State-of-the-art modeling approach 5. Recap of the take-home message
Global water cycle Source: NASA
Water between land and atmosphere Deforestation (changing from trees to crops) => Reduced evapotranspiration => Increase in runoff Evapotranspiration Evapotranspiration Runoff Runoff Wohl et al. 2012
Engaging current and predicted changes: deforestation Arc of deforestation (Current land use) By 2050, under the business-as-usual scenario, about 40% of the Amazon is predicted to be deforested (Soares-Filho et al., 2006)
Reduced rainfall predicted due to deforestation Amazonian Deforestation reduces regional Rainfall Through evapotranspiration, forests maintain atmospheric moisture that can return to land as rainfall It may alter the regional-scale and even global-scale precipitation patterns through the atmospheric circulation Climate models predict large-scale tropical deforestation causes reduction in regional-scale rainfall If current trend of deforestation continues, 12%~21% of rainfall reduction is estimated across the Amazon basin by 2050 (Spracklen et al., 2012)
Impacts of reduced precipitation to rainfall-reliant sectors Threatened Agricultural production For regions where agriculture is rain-fed, the productivity relies upon the rainfall If severe droughts happen, the food security is threatened Threatened hydroelectric power generation The depletion of water resources may give a challenge to a long-term stable supply of the electricity
Case for Brazil Hydroelectric dam in Brazil (Credit: Angela Linivo)
Case for Brazil Highly dependent upon hydropower (up to 80%) Brazilian Domestic Electricity Supply by Source 2011 From EPE
Impacts of rainfall change to hydropower Climate change and land-use change induce rainfall pattern change, and thus streamflow pattern change in rivers Changes Seasonality and Brings More Irregularity Current assumption of stationary flow in energy production models are no longer valid => Challenge to the market price estimation Changes Maximum Flow => Challenge to the energy planning such as decision of dam height Source: A.Livino (2013)
Streamflow [m3/s] Already experiencing changes: the Parana river basin Puzzling increase in stream flow from the Parana river basin since 1970s Average Max Min Itaipu Dam Hydropower plant (14,000MW) The Parana paradox: despite the reduction in rainfall, the stream flow has increased
1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 mm/year 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 m3/yr Parana paradox: Streamflow increase despite Rainfall reduction 300000 200000 Natural flow at Itaipu 10.5 % increase 100000 0 500000 400000 300000 200000 100000 0 Annual rainfall (from prescribed meteorology) 6.9 % decrease
Modeling framework How ED-BRAMS can models help improve our understanding of current changes and provide future projections? (Biosphere-regional climate model) PRECIPITATION, RUNOFFS THMB (Terrestrial Hydrology model) STREAMFLOW OF RIVERS MSUI (Energy Simulation Model)
Ecosystem Demography (ED) model Terrestrial Biosphere model Capable of representing dynamic fine-scale heterogeneity in ecosystem composition, structure and function Has been demonstrated to be ideal for studying impacts of land use changes Has been evaluated against the observations in the Amazon Early Successional Mid Successional Late Successional Each grid Solves cell is for subdivided canopy fluxes into of patches with similar water, canopy carbon structure and energy Medvigy et al. (2009)
How does a biosphere model work? Prescribed meteorology (Sheffield et al. 2006) Radiations, air temperature, precipitation, relative humidity, wind speed, and pressure (Biosphere model) Output examples: Biomass, Leaf area, Runoff
Modeled runoff reproduces the natural stream flow increase mm/year mm/year 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 250000 200000 150000 100000 50000 0 Case 1:Annual runoff with land-cover of year 1970 1970 land-cover 1969~1978 climate 18.8 % decrease 1970 land-cover 1999~2008 climate 8.5 % increase 24.4 % increase (cf) 10.5% increase 33.6 % increase in natural stream flow 250000 200000 150000 100000 50000 0 Case 2: Annual runoff with land-cover of year 2008 2008 land-cover 1969~1978 climate 11.3 % decrease 2008 land-cover 1999~2008 climate 6.9% decrease in rainfall
Parana paradox result Our findings using the ED model Model reproduces the observed phenomena! The model gives an increase in the decadal average of runoff by 8.5%, from the 1970s to 2000s. This agrees well with 10.5% increase in the actual stream flow as measured at Itaipu. Puzzle is solved! The increase in stream flow is mainly attributed to landcover change.
Ongoing and future work EDBRAMS (Biosphere-regional climate model) PRECIPITATION, RUNOFFS THMB (Terrestrial Hydrology model) STREAMFLOW OF RIVERS MSUI (Energy Simulation Model)
Tapajós river basin In the heart of the Amazon More than 8 hydropower plants are planned in next 10 years EPE, 2013
Take-home messages Water is very crucial in the processes of the atmosphere-land interactions Current and predicted deforestation alters the water cycle under anticipated climate change. Change in rainfall patterns will challenge the Brazilian energy sector as it heavily relies upon the hydroelectric power generation.
Thank you!