Atmosphere-Biosphere Interactions Prioritizing Climate Change Adaptation and Ozone Pollution Control to Minimize Impacts on Public Health and Food Security in 2050 Amos P. K. Tai ( 戴沛權 ) Earth System Science Programme, CUHK 5 th Conference on East Asia and Collaborators: Western Pacific Meteorology and Colette Heald (MIT) Climate (3 Nov 2013) Maria Val Martin (CSU)
Societal Impacts of Atmospheric Changes Human Various Human Activities consequences health Air pollution meteorology Air Pollution Climate Change Radiative forcing Agriculture Warmth � good Surface ozone Excess heat � bad � very bad Drought � bad
Earth System Modeling Framework Anthropogenic emissions, Climate and Community Earth land use (2000, 2050 ozone projections System Model RCP4.5, 2050 RCP8.5) (1.9° ×2.5° ) Surface ozone is produced in situ in the atmosphere by photochemical reaction of ozone precursors (NO x , CO, hydrocarbon). It is detrimental to human health, vegetation and crop production. chemical loss Water vapor � O 3 � low NO x Isoprene � Warming high NO x O 3 � NO x � photochemistry
CESM Ozone Projections for 2050 2000-to-2050 changes in JJA surface (MDA8) O 3 (ppbv) anthro emission only climate only biogenic emis only RCP4.5 RCP8.5 [ppb] -18 -9 0 9 18 -5 -2.5 0 2.5 5 -5 -2.5 0 2.5 5 � Anthropogenic emissions of ozone precursors are the dominant ozone driver, but climate change can partly offset (or enhance) the effects of changes in emissions.
From Atmospheric Changes to Malnutrition Anthropogenic emissions, Climate and Community Earth land use (2000, 2050 ozone projections System Model (1.9 �� 2.5 � ) RCP4.5, 2050 RCP8.5) Statistical Growing season heat relationships to and ozone exposure Changes in crop parameterize metrics for four major production: relative effects: crops: ( ) � climate and � pollution ∆ P = P 0 γ climate γ pollution − 1 Wheat Rice Distribution of per capita food consumption Maize Methodology With climate Additional and data from and pollution Soybean fraction Food and effects under- Agriculture nourished Original Organization Fraction distribution (FAO) under- nourished Minimum Dietary Energy Requirement
Climate and Pollution Effects on Crop Production RCP4.5 Pollution effect: +850 ×10 12 kcal Combined effect: -87 ×10 12 kcal Climate effect: -750 ×10 12 kcal -50 -20 -10 -5 -2 -1 0 1 2 5 10 20 50 10 6 kcal ha -1 RCP8.5 Pollution effect: -49 ×10 12 kcal Combined effect: -900 ×10 12 kcal Climate effect: -770 ×10 12 kcal
Climate and Pollution Affect Undernourishment Combined effects of 2000-to-2050 changes in climate and ozone pollution on rate of undernourishment in developing countries Rate of Based on FAO Based on FAO undernourishment 2000 data 2050 projections Baseline 18% (current) 4.1% RCP4.5 19% 4.7% RCP8.5 26% 7.7% � RCP8.5: Climate change and ozone pollution together severely threaten food security in developing regions � RCP4.5: Improvement in ozone air quality largely compensates damage of climate change; but this assumes flexible dietary habits and little barrier in international trade
Effects Vary for Different Crops and Regions Wheat Maize Soybean Rice RCP4.5 US Europe China S Asia SE Asia S America Global US RCP8.5 Europe China S Asia SE Asia S America Global -50 0 50 100 -5 5 15 25 -60 -40 -20 0 20 -40 0 20 60 Projected 2000-to-2050 % change in production Major producers 50% each 0% climate 100% climate 100% pollution 0 2 4 6 8 1 0% pollution Contribution from climate vs. pollution effect
Conclusions and Implications � Anthropogenic emissions are the dominant driver for future ozone air quality, but their effects can be substantially offset or enhanced by climate change depending on region � Globally, ozone pollution and climate change together can potentially worsen malnutrition problems, but aggressive ozone control can offset impacts of climate change � Climate adaptation will be important for climate-sensitive crops and regions: maize everywhere except in China, soybean in South America � Ozone pollution control will be important for pollution-sensitive crops and regions: wheat everywhere, rice and maize in China � Ozone pollution control have “triple” benefits of protecting public health, ensuring food security and mitigating climate change for China, where regions of high population, high ozone and croplands overlap.
Back-up Slides
Statistical Parameterization of Climate and Pollution Effects � Ozone effect: we use relative yield (RY) as function of various ozone exposure metrics found from literature � RY = 1 – a AOT40 γ pollution = RY � RY = exp[-(M12/ a ) b ]/exp[-(20/ a ) b ] 2050 RY � RY = exp[-(SUM06/ a ) b ] 2000 � Climate effect: we develop for each 1.9 °× 2.5 ° cell a constrained � RY = exp[-(W126/ a ) b ] multiple linear regression model, using 1961-2010 crop yield and meteorological data from FAO and NCEP/NCAR ( ) + β KDD KDD − KDD 5yma ( ) ln Y − ln Y 5yma = β 0 + β GDD GDD − GDD 5yma � GDD = growing degree days ( � GDD > 0) � KDD = killing degree days ( � KDD < 0) ( ) ∗ ∗ γ climate = exp β GDD ∆ GDD + β KDD ∆ KDD � � * GDD and � * KDD are the “true” effects after adjusting for confounding factors (e.g., ozone also increases with T )
Crop Response to Growing Season Temperature � We find relationships of 1961-2010 detrended annual crop yield with growing degree days (GDD) and killing degree days (KDD) for each grid: ( ) + β KDD KDD − KDD 5yma ( ) ln Y − ln Y 5yma = β 0 + β GDD GDD − GDD 5yma Constraints: � GDD � 0 and � KDD � 0 ° C T max T.SRF.daily.m ean[i, j, 152:243, 1] - 273 35 KDD T high 30 T mean 25 GDD 20 ( ) γ climate = exp β GDD ∆ GDD + β KDD ∆ KDD 15 T base 10 0 20 40 60 80 Day since planting date
Crop Response to Growing Season Temperature ( ) + β KDD KDD − KDD 5yma ( ) ln Y − ln Y 5yma = β 0 + β GDD GDD − GDD 5yma Sensitivity of maize yield to growing season T Correlation of KDD with maize 1.0 growing season mean precipitation 1.0 0.5 0.5 0.0 C-day 0.0 -0.5 � GDD -0.5 -1.0 log unit /° -1.0 4 2 KDD not only captures the 0 effect of extreme T but also -2 associated drought conditions � KDD -4 with low precipitation
Climate Effect Can be Confounded by Ozone Pollution � Ozone pollution is strongly correlated with high temperature! Therefore, part of GDD or KDD effect can be due to correlation with ozone pollution � Approach: Evaluate from + ∂ ln Y dM hourly ozone data ∗ β KDD = β KDD ⋅ (US AQS/CASTNET ∂ M d KDD and Europe EMEP for 1993-2010) Total effect Actual, non- M = ozone exposure metric confounded (e.g. AOT40); obtained by KDD effect differentiating historical O 3 - crop responses, e.g.: ∂ ln Y a = − ∂ AOT40 1 − a AOT40
Climate Effect Can be Confounded by Ozone Pollution M = α 0 + α KDD KD ′ ′ � Regression to obtain dM / d KDD = � KDD : D + ∂ ln Y dM ( ) ∗ ∗ γ climate = exp β GDD ∆ GDD + β KDD ∆ KDD ∗ β KDD = β KDD ⋅ ∂ M d KDD 40 30 Maize � � 20 ∂ ln Y dM � � � � ∂ M d KDD 10 × 100% β KDD 0 � For maize, ~7% of total KDD effect ( � KDD ) is due to correlation with ozone exposure. For wheat and soybean, it is ~20% and ~40%, respectively.
Estimation of Undernourishment Rate � � 2 ( ) � exp − ln x − µ Daily dietary energy supply (DES) 1 � � f ( x ) = or food consumption per capita � � 2 σ 2 x σ 2 π � � follows a lognormal distribution: MDER = minimum dietary MDER x energy requirement µ = ln x − 0.5 σ 2 x = mean DES With climate and pollution effect: Fraction Fraction consumed by consumed developing countries as food x = x 0 + ∆ E ⋅ 0.61 ⋅ 0.66 365 ⋅ N Food consumption per capita (kcal/person/day) Population in developing countries
Recommend
More recommend
