Expansion of Rubber (Hevea brasiliensis) and its Implications for Water and Carbon Dynamics in Montane Mainalnd Southeast Asia

Expansion of Rubber (Hevea brasiliensis) and its Implications for Water and Carbon Dynamics in Montane Mainalnd Southeast Asia Jefferson Fox Thomas Giambelluca Ryan Mudd Alan Ziegler Maoyi Huang Wen Liu Michael Nullet Qi Chen East-West Center University of Hawai i at Mānoa University of Hawai i at Mānoa National University of Singapore Battelle Pacific NW National Laboratory University of Hawai i at Mānoa University of Hawai i at Mānoa University of Hawai i at Mānoa March 2011

Jeff Fox, EWC, 2010

NR price forecasts to 2011 (It s the global market... stupid ) 3/22/2011 $5.15 Prachaya. 2009.

Traditional and non-traditional rubber-growing regions Jeff Fox, EWC, 2010

Rubber distribution in 2009 in MMSEA derived from MODIS 250m NDVI time series Jeff Fox, EWC, 2010

Jeff Fox, EWC, 2010

Rubber Estimates (2008) (non-traditional rubber areas) Country Area (ha) Northeast Thailand 348,063 (Thai Rubber 2007) Xishuangbanna, China (1950s) 334,000 (Reuters 2008) Laos (1994) 140,665 (NAFRI 2008) Cambodia 107,901 (ANRPC 2009) Northeast Myanmar (Kachin and Shan States) 68,723 (Hly Myint 2008) Vietnam 58,100 (AgroInfo 2008) Total 1,057,452 Jeff Fox, EWC, 2010

CLUE Land Cover / Land Use Simulations Overall MMSEA Results Increase Decrease Model Land Cover % 2000 % 2050 % Change Crops + Grass 12.38 14.81 2.43 Irrigated Crops 9.5 10.11 0.61 Shrubs 2.91 4.6 1.69 Deciduous Forest 12.82 15.28 2.46 (4 mil ha) (rubber) Evergreen Forest 36.58 32.85-3.73 Mixed Forest/ Mosaic 22.99 18.19-4.8

Hotspots of LCLUC in Mainland Southeast Asia (1990 to 2008) 60,000 90.00 40,000 20,000 0-20,000-40,000-60,000-80,000 F RP PA AQ PT NTR Cambodia Laos Myanmar Thailand Vietnam 80.00 70.00 60.00 50.00 40.00 30.00 20.00 10.00 0.00-10.00 F RP PA AQ PT NTR Cambodia Laos Myanmar Thailand Vietnam Total area in square kilometers Annual rate of increase in % F= forest; RP = rice/paddy; PA = protected area; AQ = Shrimp aquaculture; PT = Perennial trees; NTR = non-traditional rubber. Numbers from FAOSTAT; FAO World Forests; World Bank World Development Indicators; IUCN Protected Areas.

Prior Study Guardiola-Claramonte M, Troch Ziegler A, Giambelluca TW, Vogler JB, Nullet M (2008) Local hydrologic effects of introducing non-native vegetation in a tropical catchment. Ecohydrology 1: 13 22, DOI: 10.1002/eco.3.

Guardiola-Claramonte M, Troch Ziegler A, Giambelluca TW, Vogler JB, Nullet M (2008) Local hydrologic effects of introducing non-native vegetation in a tropical catchment. Ecohydrology 1: 13 22, DOI: 10.1002/eco.3.

Research Questions Hydrology of Hevea brasiliensis Plantations What are the hydrological and carbon consequences of conversion of land to rubber plantations in non-traditional rubber growing areas? What are the rates of ET in rubber stands, and how does it compare with other land-cover types in the region? To what extent are dry-season basin water storage affected by water use of rubber?

Tower Sites Som Sanuk CRRI Jeff Fox, EWC, 2010

Instruments and Observations Micrometeorological instruments installed at two rubber (Havea brasiliensis) plantations to measure water, energy, and carbon exchange Som Sanuk, Nong Khai, NE Thailand Trees planted in 1992 Tower installed February 2009 CRRI, Kampong Cham, Cambodia Trees planted in 2004 Tower Installed September 2009

Tower Instrumentation Cambodia Rubber Research Institute Kampong Cham, Cambodia 3 dimensional sonic anemometer (CSAT3, Campbell Scientific) Wind speed and direction (05106, RM young) 4 component radiation (NR01, Hukseflux) infrared gas analyzer (LI- 7500, Licor) PAR sensor (LI-190, Licor) Temperature and relative humidity (HMP45, Vaisala) 3 wind speed sensors for wind height profile (014A, Met one) 2 Rain gauges (TI-525, Texas Instruments) 3 air temperature sensors (TC wire, Omega) 10 biomass temperature sensors (TC wire, Omega)

Underground sensors 2 soil temperature sensors (20 cm TC probes, Campbell Scientific) 4 soil heat flux plates (HFP01, Hukseflux) 5 TDR soil moisture probes 30 cm (CS616, Campbell Scientific) 4 cm horizontal 30 cm vertical 1 m vertical 2 m vertical 3 m vertical 5 ADR soil moisture sensors (Theta Probe, DeltaT) in 2 stands (2001 and 2004) 5, 10, 20, 30, and 50 cm all horizontal

Sap flux (kg m -2 s -1 ) Collaboration with Kumagai Lab of Kyushu University for sap flow measurements 0.06 0.04 0.02 0 5/19 5/20 5/21 5/22 5/23 5/24 5/25 5/26

Som Sanuk, Thailand

Preliminary Eddy Flux Results from Som Sanuk Annual Mean Diurnal Cycles 600 500 400 300 Rubber Energy Balance Som Sanuk, NE Thailand Rnet G H LE Mean Rnet = 124 W m -2 Mean G = 2 W m -2 Mean H = 27 W m -2 Mean ET = 87 W m -2 ECE = 8 W m -2 ECR = 91% 200 100 0-100 0:00 3:00 6:00 9:00 12:00 15:00 18:00 21:00 0:00

Preliminary Eddy Flux Results from Som Sanuk 180 160 140 120 100 80 60 40 20 0 Rubber Energy Balance Som Sanuk, NE Thailand Rnet G H LE -20 Feb-09 Apr-09 Jun-09 Aug-09 Oct-09 Dec-09 Feb-10 Apr-10

Seasonal Cycle Related to Leaf Area 180 160 140 120 100 80 60 40 20 0 Rubber Energy Balance Som Sanuk, NE Thailand Rnet G H LE -20 Feb-09 Apr-09 Jun-09 Aug-09 Oct-09 Dec-09 Feb-10 Apr-10 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 17 yr old 12 yr old 6 yr old Leaf Area Index 0.0 Feb-09 Apr-09 Jun-09 Aug-09 Oct-09 Dec-09 Feb-10 Apr-10

NEE Preliminary Eddy Flux Results from Som Sanuk: Mean Diurnal NEE Cycles NEE ( mol m -2 s -1 ) 10 5 Jan 0-5 -10-15 -20-25 -30 Oct Sep Aug Dec Nov Jul 0 6 12 18 24 30 36 42 48 30-minute Timesteps May Jun

NEE Preliminary Eddy Flux Results from Som Sanuk: Mean Diurnal NEE Cycles NEE ( mol m -2 s -1 ) 10 5 Feb 09 Mar 09 Apr 09 May 09 Jun 09 Jul 09 Aug 09 Sep 09 Oct 09 Noc 09 Dec 09 Jan 10 Feb 10 Mar 10 0-5 -10-15 -20-25 -30

Mean Annual ET: 1100 mm Site Land Cover LE:Rnet Borneo Tropical rainforest 0.89 Amazon Tropical rainforest 0.86 Som Sanuk Rubber plantation 0.70 Chiang Mai Hill evergreen 0.60 Mato Grosso Transitional forest 0.66

Weimin Ju,Fangmin Zhang, Jingming Chen, Shuanghe Shen, Shaoqiang Wang, Guirui Yu, Xinquan Zhao, Shijie Han, J. Asanuma (2010) Trends of evapotranspiration in East Asia from 1982 to 2006 simulated using a remote sensing-driven ecological model. Presented at HESSS2, 22 June 2010, Tokyo.

Ecosystem Model Must Incorporate Vegetation Dynamics Bonan (2008, Science 320: 1444-1449)

Simulation of forest biomass density in Mainland Southeast Asia Initialization 100 years 200 years 300 years

Simulations of the forest biomass density with the incorporation of land use/cover change Beginning 100 years 200 years 300 years 400 years 500 years

Preliminary verification of carbon Carbon map of Laos 2010 Modeled carbon density after 500 years of land use

Summary Rubber ET may be higher than forest ET Seasonal cycle significantly changed with lower Sep-Jan ET and higher Mar-Jul ET Annual Carbon flux cycle strongly influenced by the phenology of rubber Question: What effect does the altered annual cycle have on basin water storage and river discharge at the start of rain season? How does the annual and long-term carbon budget of rubber plantations compare with those of land covers that rubber is replacing? Further work: Continued monitoring-both sites Integrate EC and Sapflux observations Leaf level measurements Installation of carbon profile monitoring equipment Carbon stocks and budget

Thank You

CRRI