On the Use of Agricultural System Models for Exploring Technological Innovations Across Scales in Africa: A Critical Review
Reimund Rötter,Fanou Sehomi,Jukka Höhn,Jarkko Niemi,Marrit Van den Berg,Reimund RRtter,Jukka HHhn
DOI: https://doi.org/10.2139/ssrn.2818934
2016-01-01
SSRN Electronic Journal
Abstract:The major challenge of the 21st century is to achieve food security under, roughly, a doubling in food demand by 2050 compared to present, and producing the additional food under marked shifts in climatic risks and with environmentally sound farming practices. Sustainable intensification of agricultural production is required that meets the dual goal of improved environmental sustainability and economic efficiency. Ex ante evaluation of technological innovations to support agricultural production and food security taking into account the various future risks can substantially contribute to achieve this. Here we perceive technological innovations as new or improved agro-technologies and – management practices, such as new breeds, integrated soil fertility practices or labour-saving technologies meeting the goals of sustainable intensification. In this report we present results from three systematic reviews: one on the use of biophysical modelling, the second and third on the use of bio-economic modelling at farm scale and agro-economic modelling at higher aggregation levels, for ex ante evaluation of the effects of (agro-) Technological Innovations (ag-TIs) on sustainable agriculture and food security indicators. To this end, we searched the SCOPUS database for journal articles published between 1996 and 2015. We considered modelling studies at different spatial scales with particular attention to local to national scale studies for the twelve PARI focal countries in Africa. But we also included studies for all other African countries as well as a few studies at supra-national/continental scale. Both, “quick wins” as well as long term benefits from ag-TIs were of interest. The various ag-TIs were furthermore grouped into four classes: (1) water/soil moisture (2) soil nutrients/conservation (3) crop/cropping system, (4) other ag-TIs or (5) combinations of 1 to 4. For each paper, we tried to identify the primary ag-TI analysed, and if there was equal emphasis to more than one, we classified them as combinations. It should be borne in mind that there is some subjectivity in classifying the papers in this way. Results. After various steps of refining “search strings”, screening on relevance and supplementing databases from additional sources, we found 140 relevant biophysical modelling studies, whereby coverage of sub-regions and ag-TIs varied markedly. Most studies were found for East and West Africa, followed by Southern Africa; hardly anything was found for Northern and Middle Africa. A similar pattern appeared for the integrated agro-economic modelling studies at farm scale, for which we found 40 relevant ones. Agro-economic modelling studies at higher aggregation levels showed a somewhat different pattern – and more generally contained little detail on technological innovations. Regarding the share of different primary agro-technologies explored in the biophysical studies we found 45 on crop management, 35 on combined agro-technologies, 31 on soil nutrient management and conservation, 23 on water/soil moisture management, and 6 on other technologies. We found similar shares among the various agro-technology groups for the integrated agro-economic modelling studies at farm scale. Looking at the outcomes from ex ante evaluations we found that many studies are (mostly) positive on effects of single and “conventional” ag-TIs. The majority of biophysical studies is performed at “field scale” and focuses on the effects on productivity (sometimes yield stability); many of these studies were performed in climate variability and change/adaptation research context. Most agro-economic modelling studies that look specifically at ex ante evaluations of ag-TIs are performed at farm or regional (sub-national) scales. While the number of biophysically oriented studies has grown exponentially over the considered period 1996-2015, this is not the case for the agro-economic modelling studies. Looking in more detail at the twelve focal countries of PARI (=Programme of Accompanying Research on Agricultural Innovations) we also find an unbalanced distribution, with most studies found in Kenya, Ethiopia, Mali and Ghana (biophysical modelling studies), and respectively in Kenya and Uganda (agro-economic modelling studies), whereas nothing or little was found for both types of studies in Togo, Zambia and Nigeria. Very few of the biophysically-oriented studies include other information than effects on crop yields, and there are few studies for both biophysical and agro-economic modelling that comprise multi-scale or higher scale analyses; if multi-scale, there are more studies that scale up from field/farm to regional/sub-national level than from field/farm to nation scale or beyond. There is definitely a need to overcome the lack of meaningful integrated multi-scale modelling along the lines proposed in chapters 5-6 of this report. Moreover, less than half of all integrated/agro-economic modelling studies at farm scale explicitly address risk – another clear shortcoming, which requires attention by the research community. A more general conclusion is that there is no application yet of true transdisciplinary research approaches in practice. Hence, there is need for participatory, collaborative (cross-sectoral) and combined modelling approaches with adequate stakeholder involvement throughout the research process. In this respect, some lessons might be learned from pioneering work conducted in Asia and Europe.