Decadal Satellite Observations and the Myth of Malawian Farm Input Subsidy Programme
0 likes475 views
The document discusses the Malawian Farm Input Subsidy Programme (FISP), challenging its perceived success in doubling agricultural production and highlighting that the 'Malawian production miracle' is largely a myth. It examines various factors affecting agricultural land suitability, climate variability, and production trends in Malawi, indicating that many farms suffer from degraded soil organic matter and inconsistent crop yields. The text advocates for improved targeting of biophysical solutions and suggests that traditional development approaches may not adequately address the complexities of agricultural challenges in the context of climate change.
Decadal Satellite Observations and the Myth of Malawian Farm Input Subsidy Programme
- 1. 9/3/15 1 Joseph P. Messina Ph.D. Associate Dean for Research and Professor College of Social Science, Center for Global Change and Earth Observations, the Department of Geography, Michigan State University. jpm@msu.edu 517-‐432-‐3436 Decadal Satellite Observations and the Myth of Malawian Farm Input Subsidy Programme Farm Input Subsidy Program (FISP) in Malawi • This widely discussed subsidy program has been praised in the press as “brilliant” leading to “(doubled production) within one harvest season” (J. Sachs, NYT April 19, 2012) • All popular press sources referred to government statistics regarding production yields as having significantly increased and being positive in the near term for small-holder farmers. • As Predro Sanchez commented (Nature: 2015) “in spite of criticisms by donor agencies and academics, the seed and fertilizer subsidies provided food security to millions of Malawians.” • This optimistic assessment of potential for an African Green Revolution must be tempered by the fact that the Malawian production miracle is largely a myth. http://www.faceofmalawi.com/2014/10/donors-‐demand-‐ full-‐details-‐of-‐fisp/
- 2. 9/3/15 2 An Accidental Discovery and how I started this process. • Perennial grains to improve resilience to climate change. • Malawi, Mali, Ghana, Tanzania • Pigeon Pea in Malawi • Where are the marginal lands? 1. Agricultural lands? 2. Production trends 3. Inter-‐ vs. inter-‐annual trends Crop Models? • DSSAT, SALUS, APSIM, … • Scale • Data uncertainty
- 3. 9/3/15 3 Yield? • Yield = f(edaphic * climate * crop * social) • Multiscalarprocesses • Soils are local but also a product of watershed dynamics • Climate is global but also a local physical process • Social is endogenous (labor, capital,…) and exogenous (FISP, market prices, trade, …) • Agronomy is local but also responsive across scale • Food security scales from local to national. Global Change Biology. doi: 10.1111/gcb.12838 Uncertainty & Social vs. Biophysical Drivers • CRU trend versus ERA40 trend in rainfall. • Blue=wetter trend, • Orange = drying trend. • Different spatial scales and process-based methods that use THE SAME STATION DATA lead to drastically different outcomes based mainly on scale (i.e. where blue and orange overlap). • So, How do we fix this problem?
- 4. 9/3/15 4 Where are the marginal agricultural lands? Mozambique Tanzania Zimbabwe Zambia Agricultural Land Classification Disagreement in Malawi Coverage of Agricultural Land FAO 2010 & IFPRI 2002 - 35% IFPRI 2002 Only - 26% FAO 2010 Only - 16% Non Ag 0 10050 Kilometers • What does this even mean? • Standard LULC products answer the wrong question (How much?) and they try to minimize overall classification error • I need to minimize errors of commission. – a very reliable sample… • Solution: use all LULC products
- 5. 9/3/15 5 RS based LAI vs. SALUS Crop Model But where is the FISP bump??? Soils and climate and other things we need to distinguish • Marginal lands – relative terms • Soils – need to extract soil drivers • Climate – inter and intra annual variability • Scale – local heterogeneity is likely a social process
- 6. 9/3/15 6 Factors: 1. Slope 2. Soil erosion hazard 3. Soil bulk density 4. Soil organic matter 5. Soil cation exchange capacity 6. Soil texture 7. Soil pH 8. Soil drainage 9. Soil depth Methods: Average of 1. Geometric mean 2. Rabia 3. Square root 4. Storie 5. Weighted sum Malawi Agricultural Land Suitability Categories Area (ha) Percentage Highly suitable 915431 7.8 Moderately suitable 2458882 20.8 Marginal suitable 2379508 20.1 Suitable land total 5753821 48.7 Poorly suitable 1081160 9.2 Permanently unsuitable 2496676 21.1 Unsuitable land total 3577836 30.3 Land subtotal 9331657 79.0 Water 2479429 21.0 Malawi total area 11811086 100.0
- 7. 9/3/15 7 Optimal Pigeon Pea • Niche generation • Extensive data – but a manageable process • Organized by management unit to facilitate scaling and adoption • But, • Sensitivity to climate? • Variability across scales? • This is not substantially different than a well parameterized crop model
- 8. 9/3/15 8 Production trends • Relative terms • Malawi specific scale • Sensitive • High inter-‐annual variability • Missing? • Scaled trends • Sources of the variability
- 9. 9/3/15 9 Rainfall and Productivity Trends on Agricultural Land in Malawi
- 10. 9/3/15 10 Now what? • We can improve targeting. • Biophysical solutions need to target biophysical problems • Social solutions … • Adoption is cultural • Next steps? Climate & Changing Seasons
- 11. 9/3/15 11 FISP comments • There are multiple lines of evidence that on many farms soil organic matter status has degraded to a level that no longer support’s maize growth or responsiveness to fertilizer. • No clear trends of improvement (Dorward et al., 2010a). • The incremental production estimates are, however, considerably lower than those implicit in the national crop estimates for maize production, with much lower variation. (Dorwardet al., 2010a). • Annual changes in maize prices also suggest that post-‐subsidy maize supplies have been lower than suggestedby the national crop estimates. (Dorwardand Chirwa, 2011) • “It is widely believed that the 2007 Malawi harvest was overestimated by at least 25%. If the government had been able to produce a more accurate estimate of crop production, it might not have arranged to export maize, which in turn might have avoided the huge price surge in late 2007/early 2008 which caused great hardship for maize buying households.” (Jayne, 2008)
- 12. 9/3/15 12 The development orthodoxy on agricultural • Intensification (all the time) • Better varieties • Better supply chains • Better subsidies • Scaling matters • A function of targeting • The Climate Change Quandary • The new varieties may not be appropriate • Tools • Adoption potential • Unintended Consequences • Disease – livestock, plant, and human • Climate and Ag feedbacks • Big Data solutions will help solve some of the scaling, climate, and targeting challenges faced by the development community.
- 13. 9/3/15 13 Funding Provided by: • The Bill & Melinda Gates Foundation • USAID – Global Development Lab & GCFSI Thank You / Questions Reference: Decadal Satellite Observations and the Myth of Malawian Farm Input Subsidy Programme. 2015. Messina, J. Peter, B. Li, G. DeVisser, M. Snapp, S. Moore, N. Nejadhashemi, P. Putting Perennial crops to work in practice: Pigeonpeas and Sorghum. Bamako, Mali