Food security and adaptation in the context of potential CSA

Download Report

Transcript Food security and adaptation in the context of potential CSA

Food security and adaptation in the
context of potential CSA practices in
Zambia
Aslihan Arslan
(Co-authors: Nancy McCarthy, Leslie Lipper, Solomon Asfaw, Andrea Cattaneo and
Misael Kokwe)
1st Africa Congress on Conservation Agriculture
19.03.2014
Lusaka, Zambia
Outline
•
•
•
•
•
•
•
CSA & CA
Background
Data sources
Climate variables
Descriptive stats
Results
Conclusions
Climate Smart Agriculture
FAO CSA 2010 definition:
Agriculture that sustainably increases productivity, resilience (adaptation),
reduces/removes GHGs (mitigation), and enhances achievements of national food
security and development goals.
CSA = CA?
• CSA:
• is an approach to achieve agricultural development
under climate change
• CA:
• has the potential to contribute to CSA pillars
• different impacts in different locations & experimental
vs. farmer plots
• barriers to adoption (e.g. opp cost of residue, time
delay)
• needs to be studied under farmer conditions & climate
change lens
Questions Addressed
1. What are the impacts of CSA practices on maize yields per
hectare in Zambia?
2. What are the impacts of CSA practices on the probability of
very low yields and on the yield shortfall?
Practices Studied:
1. Minimum Soil Disturbance (MSD)
2. Crop Rotation (CR)
3. Legume Intercropping (LEGINT)
4. Inorganic Fertilizer Use (INOF)
5. Improved Maize Seeds (IMPS)
CSA??
Data Sources 1
• RILS 2004 and 2008: supplemental surveys
(CSO/FSRP) to the annual post-harvest surveys
(PHS)
– Both nationally representative
– Around 4,000 households interviewed in both years
– 4,138 & 4,354 maize plots in 1st and 2nd rounds
– Econometric analyses of productivity and probability
of low production controlling for a large set of
relevant socio-economic, climate and agro-ecological
variables
RILS Enumeration Areas & AER
Data Sources 2
• Rainfall (1983-2012): Dekadal (10 days) rainfall
data from Africa Rainfall Climatology v2 (ARC2) of
the National Oceanic and Atmospheric
Administration’s Climate Prediction Center (NOAACPC)
• Temperature (1989-2010): Dekadal avg, min &
max temperatures of the European Centre for
Medium-Range Weather Forecasts (ECMWF)
• Soil: Soil nutrient availability and soil pH levels from
the Harmonized World Soil Database (HWSD)
Climate Variables
• Rainfall:
1. Growing Season Total (and its square)
2. Onset of the rainy season: 2 dekads of >=50mm rainfall
after October 1.
3. Dry spells: # dekads with <20mm rain during
germination&ripening
4. False onset: 1 dekad with <20mm rain after the onset
• Temperature:
1. Growing season average
2. Growing season max
3. Indicator if Tmax=28 degrees
References:
Tadross et al. 2009. “Growing-season rainfall and scenarios of future change in southeast Africa:
implications for cultivating maize. “ Climate Research 40: 147-161.
Thornton P., Cramer L. (eds.) 2012. “Impacts of climate change on the agricultural and aquatic
systems and natural resources within the CGIAR’s mandate.” CCAFS Working Paper 23.
Maize Yields by AER & Year
Maize Yields by AER
2004
2008
0
.0002
.0004
.0006
.0008
Maize Yields by AER
0
2000
4000
6000 0
2000
x
AER I
AER IIb
Graphs by year
AER IIa
AER III
4000
6000
Season total rainfall by AER & year
Season Rainfall by AER
2004
2008
0
.005
.01
Season Rainfall by AER
500
1000
1500
500
1000
x
AER I
AER IIb
Graphs by year
AER IIa
AER III
1500
Average Temperature by AER & year
Season Avg. Temp. by AER
2004
2008
0
.5
1
1.5
Season Avg. Temp. by AER
20
22
24
26
20
22
x
AER I
AER IIb
Graphs by year
AER IIa
AER III
24
26
Max Temperature by AER & year
Season Max. Temp. by AER
2004
2008
0
.5
1
1.5
Season Max. Temp. by AER
24
26
28
30
24
26
x
AER I
AER IIb
Graphs by year
AER IIa
AER III
28
30
CoV of Rainfall & Onset by AER
50
100
kdensity onset_cov8312
40
30
20
0
0
10
CoV of Rain onset (1983-2012) by AER
150
CoV of Rainfall by AER
.1
.15
.2
x
AER I
AER IIb
.25
.3
.01
.02
.03
.04
.05
x
AER IIa
AER III
AER I
AER IIb
AER IIa
AER III
.06
Shares of maize plots under
each practice
Year
2004
2008 Total
MSD
0.030*** 0.043***
0.037
CR
0.239*** 0.361***
0.301
LEGINT
0.047*** 0.029***
0.038
INOF
0.374
0.391
0.382
HYBM
0.436*** 0.476***
0.457
MSD+CR
0.009*** 0.021***
0.015
MSD+LEGINT
0.001
0.001
0.001
MSD+INOF
0.010
0.008
0.009
MSD+HYBM
0.010
0.010
0.010
CR+LEGINT
0.007
0.007
0.007
CR+INOF
0.087*** 0.143***
0.115
CR+HYBM
0.079*** 0.146***
0.113
LEGINT+INOF
0.011**
0.007**
0.009
LEGINT+HYBM
0.014*** 0.006***
0.010
INOF+HYBM
0.217*** 0.259***
0.238
CR+INOF
0.052*** 0.098***
0.075
LEGINT+INOF+HYBM 0.007*** 0.003***
0.005
* significant at 10%; ** significant at 5%; *** significant at 1%
Maize yields by practice & year
Average maize yields (kg./ha) by practice and year
2004
MSD
CR
LEGINT
INOF
HYBM
No
1,580
1,538
1,576
1,320
1,417
2008
Yes
1,495
1,703
1,619
2,011
1,786
No
1,551
1,513
1,538
1,206
1,229
Yes
1,317
1,589
1,629
2,060
1,884
Econometric Analyses
The methodology we use…
• Avoids confounding factors that affect average
yield comparisons (e.g. farmer characteristics,
plot characteristics, labor availability, other input
use)
• Helps us identify the average impact of a practice
on yields and probability of very low production
• Interaction terms between climate variables and
practices help us identify how the average
impacts vary with climatic conditions
Summary of robust findings
Yield
MSD
CR
LEGINT
INOF
IMPSEED
CR*CoV Rain
INOF*CoV Rain
INOF*False onset
IMPS*False onset
IMPS*tmax ≥28°C
Fertilizer on time
Rainfall
Max temp ≥ 28°C
+
+
+
+
+
+
+
p(low
yield)
Yield
shortfall
-
-
+
+
+
+
+
+
+
-
Conclusions- yield effects
• Climatic shock variables significantly change the
impacts of practices
• Rainfall variability drives yield effects: In high
variability areas…
• Crop rotation has positive effects
• Inorganic fertilizer & hybrids not effective
• Legume intercropping has robust yield impacts
• No significant impact of minimum soil
disturbance on yield outcomes
• Timely fertilizer delivery most important
Broader implications
• Data used are from years with limited rainfall
stress
• Our analysis shows that some climate related
variables determine which practices will yield
best results
• Taking climate variables into consideration in
developing strategies to support agricultural
productivity increases is essential.
• Our results suggest SLM/CA practices could play
an important role in responding to CC.
THANK YOU!
APPENDIX
Independent variables used in empirical models
Variables
Age of household head
Education (average)
# of adults (age>=15)
Share of ill adults
Female headed
Total maize area (ha)
Wealth index
# of oxen owned
Organic fertilizer applied
# of weedings applied
Tilled before rainy season
Policy Variables
ASP Dummy
Had fertilizer on time
Geo-referenced Variables
Growing season rainfall (100mm.)
CoV of growing season rainfall (1983-2012)
False onset of rainy season
Growing season avg. temperature (°C)
Growing season max. temperature ≥ 28°C
Moderate nutrient constraint
Severe/very severe nutrient constraint
Average soil pH
Observations (# maize plots)
2004
49.50
5.23
4.58
0.07
0.21
1.09
0.21
0.78
0.12
1.72
0.37
0.50
0.29
8.62
0.20
0.63
21.96
0.14
0.35
0.35
5.59
4,138
2008
52.48
5.47
3.91
0.02
0.21
1.52
0.18
1.18
0.12
1.70
0.33
Signif.
***
***
***
***
***
***
***
***
0.53 **
0.34 ***
8.19
0.21
0.19
22.27
0.18
0.34
0.34
5.61
4,354
***
***
***
***
***
Fertilizer timeliness by province
& land size
Province
Central
Copperbelt
Eastern
Luapula
Lusaka
Northern
Northwestern
Southern
Western
Total
2004
0.43
0.45
0.28
0.33
0.45
0.31
0.11
0.29
0.05
0.29
2008
0.54
0.49
0.30
0.16
0.54
0.35
0.20
0.31
0.03
0.32
Land size
<=1.5ha
1.5-2.5ha
2.5-5ha
5-20ha
>20ha
Total
2004
0.22
0.28
0.34
0.46
0.55
0.29
2008
0.25
0.30
0.36
0.46
0.53
0.32
Further EPIC Work
• Similar analyses on the impacts of sustainable land
management practices on yields, incomes and food
security in Tanzania, Malawi, Uganda, Niger, Nigeria,
Ethiopia with detailed climate data
• Analyses of climatic shocks and welfare in these
countries
• Work with ministries of agriculture in Malawi &
Zambia to design CSA policies
• Support to MS and PhD students to work on CSA
• Investment proposals for CSA (potentially targeting
GCF/GEF for funding)