|
Production : Production Systems : Response
Response
Crop production systems include components of weather, soils, crop varieties, and human choices and actions that determine, for example, the spatial and temporal mix of crops and production inputs. A unique feature of the HarvestChoice spatial evaluation framework is its inbuilt capacity to assess the potential impacts of change on the likely future performance of production systems. This flexible and dynamic capacity is enabled through the use of a range of crop growth and cropping system simulation models. Such models mathematically describe the growth and development of crops interacting with the environment and human management of the production process [1]. With appropriate levels of calibration and validation against empirical data, crop models can be used to respond to "what if...?" questions about the potential impacts of changes in technology, management practices and environmental drivers such as pest and disease incidence and climate change. HarvestChoice has provided the analytical framework within which a number of cross-discipline and cross-institution teams of specialists are providing the necessary levels of expertise to design, implement and validate a robust performance response capacity.
Traditionally, crop models have been used to simulate crop growth in a single field (a "homogeneous management unit"). However, HarvestChoice is scaling up the application of crop models across much broader geographic extents e.g. across sub-Saharan Africa and globally [for other examples see 2,3]. One unique feature of the effort being made by HarvestChoice is to characterize contemporary patterns of cropping systems from which basis the impacts of change will be assessed. This is being done through extensive (on-going) analysis of nationally representative production censuses and surveys for sub-Saharan Africa and South Asia. By benchmarking the crop modeling platform around a richly-disaggregated perspective of existing production systems, HarvestChoice is equipping itself to provide significantly more reliable assessments of the potential economic impacts of a wider range of specific interventions.
The performance of crop models may be compromised if the environmental or production conditions of interest greatly differ from those used during model development and evaluation. In new use contexts, therefore, adjustments in model structure and/or parameters may be necessary. As part of its work with crop growth models, HarvestChoice is designing and implementing processes to assess and improve the reliability of the models for the primary geographies and cropping systems of interest. Such validation will be critical in gaining the confidence of scientists, analysts, investors, and policymakers.
Crop models can provide estimates of change in important system performance indicators other than crop yield including; changes in total above- and below-ground biomass (valuable for fodder and biofuel assessments), changes in carbon and nitrogen stocks and flows, and changes in consumptive use of water and nutrients. This greatly enriches the range of potential "what if...?" questions the HarvestChoice evaluation framework can assess.
 back to top
Potential Yields
The description of HarvestChoice's cropping system modeling capacity is divided into two parts. Here we describe activities and outputs around its use in assessing the performance of existing production systems, noting that on-going, parallel work on will increase the specificity of our production system definition in particular locales and socioeconomic contexts. The use of this increasingly validated cropping system platform to answer specific "what if... ?" questions and generate a range of potential performance responses to change is described under the Biophysical Experiments sub-page in the Evaluation section.
At present, we use only three aggregate definitions of production systems' characterization. For each of the three systems (i.e., subsistence-oriented, rainfed, low-med input; market-oriented, rainfed, med-high input; and market-oriented, irrigated, med-high input), a crop growth model is used to estimate the yield potential under best practices with existing crop varieties and management assumptions. Currently this does not include the impacts of biotic constraints or related mitigation practices, but these factors will be incorporated as the parallel HarvestChoice pest and disease mapping and damage work proceeds. That integration will involve assessing potential yield losses through pest and disease damage pathways that can be represented in the crop simulation framework (e.g. pest damage to leaves, stems or roots, as well as weeds that can be modeled as competitive crops).
The following examples show maize baseline yield potentials estimated using regionally characterized maize farming systems at 30' grid-cells across sub-Saharan Africa.
|
LOW INPUT
(subsistence-oriented) |
MEDIUM INPUT
(market-oriented) |
HIGH INPUT
(market-oriented) |
Nutrient management |
No supplementary plant nutrients
(fallow/rotations) |
Fertilizer application at
20 kg[N]/ha rate |
Fertilizer application at
50 kg[N]/ha rate |
Water management |
No supplementary irrigation or soil moisture conservation practices |
Irrigate when soil moisture is completely depleted
(wilting point) |
Irrigate when soil moisture is 50% depleted
(midway between field capacity and wilting point) |
Germplasm |
Traditional variety
(long maturity) |
Improved variety
(medium maturity) |
Improved variety
(medium maturity) |
Simulated maize yield potential by production system
(assuming best practices) |
 |
 |
 |
Aggregation
by country |
 |
|
|
Aggregation
by agroecological zone |
 |
|
|
back to top
Yield Gap Analysis Tool
When aggregated, the difference between the estimated actual crop yield (crop allocation) and the simulated potential crop yield under best practices provides a valuable, spatially explicit layer of information that can be used in a regionwide yield-constraint and intervention targeting analyses. Yield-gap analysis is based on the notion that there are a range of factors that determine the difference between the potential yield of a crop at any given location and the yield actually realized under a specific set of crop management conditions. Better understanding of the magnitude of that gap and the relative attributes of different yield limiting factors can provide rich insights into the potential economic benefits of different yield loss mitigation strategies.
back to top
Yield Simulation Mapping Tool
HarvestChoice uses four different crop models to answer to different types of questions at different scales. This strategy of model diversity is being pursued to reflect that individual models have different strengths and weaknesses in evaluating
different types of crop systems and interventions. Following maps show the preliminary results of global crop production simulated for 13 crops in 36 growing conditions at 1-degree grid cells using WOFOST model. This global scale analysis is useful to develop qualitative indices to compare crop responses across regions.
back to top
|
