Julian R - Assessing the Impacts of Climate Change on SSAn and SEAn Agriculture (PhD transfer)

Assessing the impacts of climate change on Sub-Saharan Africa and South-East Asian

agricultureJulian Ramirez-VillegasClimate Impacts Group, ICAS

(c) Neil Palmer (CIAT)

Contents

• Project title and supervision• Background and rationale• Objectives• Research topics• To-date results: aspects of

climate relevant to crop production modelling

• Other results

Project title and supervision

• Title: Informing the adaptation of agricultural systems in Africa and Asia to progressive climate change over the coming decades

• Supervision– Andy Challinor (principal @SEE, Leeds)– Andy Jarvis (external @CIAT, Colombia)– Doug Parker (nominal @SEE, Leeds)

• Assessment– Peter Knippertz (RSG member @SEE, Leeds)

(c) Neil Palmer (CIAT)

Background and rationale• Agriculture contribution to GDP is between 3 to 61% in

developed countries (World Bank, 2008)

Background and rationale• By 2100, novel climates could happen in 10-48% of the earth

(Williams et al. 2007)• Climate change is predicted to decrease agricultural yields (many

authors), with major impacts in the DW (many authors)

Source: Lobell et al., 2008

Background and rationale

• Hence, three challenges are expected– Adapting agriculture to future stresses– Meeting the future food demand– Mitigating GHG emissions

• Requiring some R&D to be done:– Generating base information for I.A.– Assessing impacts on agriculture at different scales– Development of adaptation strategies– Test and transfer of these strategies

General objectives

• Assess the impact of climate change on agriculture in 12 countries of Africa and South-East Asia for a set of important crop production systems

• Produce a set of recommendations on how to adapt these systems to avoid crop yield drops in the future.

Workflow

1

2

3

Research overview

Research focus regions

Research topics

• 1. Assessing relevant climate data for agricultural applications– Climate datasets screening– Assessment of available and relevant data• Quality• Uncertainties

– Analysis of GCM skill (CMIP3 and CMIP5) on crop-growth related variables

– Future climate projections uncertainty quantification

Research topics

• 2. Assessing impacts of climate change on agricultural production– Selection of crops based on regional relevance– Development, calibration and evaluation of

EcoCrop for the target crops– Knowledge gap filling, calibration and evaluation

of GLAM for the target crops– Predicting current, future yields and suitability– Detecting vulnerable areas and systems

Research topics: aspects of climate relevant to crop production modelling

• Photosynthesis is the fundamental process of interest

Storage in organs

Research topics: aspects of climate relevant to crop production modelling

• And is affected by a number of factors

Source: Isdo et al., 1995 (sour orange trees) Source: Bates, 2002

Temperature and CO2 Avail. water and solar radiation

EcoCropRamirez et al. (accepted for publication)

It evaluates on monthly basis if there are adequate climatic conditions within a growing season for temperature and precipitation… …and calculates the climatic suitability of the

resulting interaction between rainfall and temperature…

• For assessing crop climatic suitability…

GLAMChallinor et al. (2004)

• Designed at climate model scale to capitalise on known large-scale relationships between climate and crop yield, thus avoiding over-parameterisation.

Uses grid-scaled agricultural statistics to simulate yields

To simulate yields at climate model scale

Large-area models are able to reproduce large-scale historical yield responses to climate and inter-annual variability

Observed peanut yields (kg/ha) Rate of simulated to observed yields

Research topics• 3. Adapting agriculture to climate

change through crop management and genetic adjustments– Genetic level modifications

• Drought and waterlogging tolerance• Heat and cold tolerance

– Management strategies• Tillage as an improvement of soil phys.

Characteristics• Shading as in albedo or extinction

coefficient

– Determining best combinations of Management-Genetic improvements

Research topics: useful- and uniqueness

• Many studies focusing on the developing world• Many using process-based models• Few with broad coverage• Few with adaptation-focus• Few with proper uncertainty quantification (avg.

num. future scenarios is 3)• Few with solid background on climate science• Few using niche-based approaches• None using niche-based AND process-based

models at the same time…

Results to date

Results: aspects of climate relevant to crop production modelling

• So, accurate measurements are required

0.4

0.4

0.4

0.4

0.8

0.8

1.2

1.2

1.6

2.8

2.8

3.6

4.0

7.7

10.5

10.9

50.4

0.0 20.0 40.0 60.0

GHCN

GPCC

GPCP

PRISM

ATEAM

VEMAP

ARTES

GSOD

MARS

CRU-CL

WorldClim

Satellite

RCM

Other

GCM

CRU-TS

Weather station

Percent of studies

0.8

0.8

4.8

5.6

8.7

17.5

19.0

42.9

0.0 20.0 40.0 60.0

ARPEGE

Unclear

SC Variables

WG GCM

PS GCM

SD GCM

RCM

AI GCM

Percent of future-climate related studies

Common sources of present-day climate data

Common sources of future climate data

Field measurements

AND

Climatologically robust future

prediction methods

Source: Ramirez and Challinor, in prep.

*on the basis of 205 peer-reviewedpapers

Results: aspects of climate relevant to crop production modelling

• Complexity of the systemHumid west Africa

Sahel

IGP

East African highlands

East Africa arid lands

+++Constraints to data and uncertainty quantification

Results: aspects of climate relevant to crop production modelling

• Global climate model skill (IPCC 4AR)

Source: Ramirez and Challinor, in prep.

Mea

n te

mpe

ratu

reD

iurn

al te

mpe

ratu

re ra

nge

Rain

fall

Annual December-February June-July-August

Results: aspects of climate relevant to crop production modelling

• Climate model skill (CMIP3)1961-1990 Rainfall 1961-1990 Temperature

Source: Ramirez and Challinor, in prep.

GISS-MODEL-EH

NCAR-CCSM3.0

R-square (observed vs. climate m

odel) - RAINFALL

Range: 1-347Mean: 41 ± 63 points/gridcell

Range: 1-80Mean: 10 ± 12 points/gridcell

Mean bias rate (observed vs. clim

ate model) - RAIN

FALLGISS-MODEL-EH

NCAR-CCSM3.0

Rain is mostly overestimated

Other results

• Development and evaluation of the EcoCrop model and a case-study with sorghum (Crop-climate ensembles Special Issue in AFM)

• Crop data being organised in a database with stable design

Future plans• Year 2:– Finalise data collection (agricultural statistics and crop-

presence)– Gather and asses CMIP5 climate model outputs– Physiology knowledge gap filling in GLAM– Crop model (EcoCrop) calibration and evaluation for other

crops– Crop model (GLAM) calibration and evaluation

• Year 3:– Queries to experts and literature on priority traits and

realistic ranges– Modifications to crop models to test realistic ranges of

genetic and/or management strategies– Assessing adaptation options

Thanks