Crop area and crop production estimation for small farms

  1. Deutsch nicht verfügbar
  2. English

SALSA developed an estimation of the area occupied by the key crop products, as well as, its crop production in small farms for each region. Crop area and crop production estimation for small farms offers an average for five years. The predictability and reliability of the data is very important, SALSA has managed to produce fairly accurate information using a complex method. This product in addition to another output, a crop type map, covers over 21 reference regions (case study areas) of the project (20 of them in the EU). These outputs serve the overarching aim of SALSA to support the assessment of small farms’ role in contributing to food production and food security. More specifically, SALSA aimed to demonstrate the benefit of remote sensing technology in providing accurate and timely information on the crop types, area extent, and yield estimates. Such information is crucial to objectively quantify the crop production capabilities of small farms.

For the estimation of the crop area, the area covered by each key crop product cultivated by small farms in each region was extracted from the regional statistics and regressed against the unbiased crop area (plots < 5ha) estimated by Sentinel using linear regression. The results of crop area estimation show that crop area obtained from Sentinel data can be used with confidence, in particular for those regions where this information is absent from official statistics. However, given this is a snapshot picture, the point validation may be needed in the future, to account for climate and other disturbance factors that may have affected yields.

For the estimation of crop production, the project combined unbiased crop area estimations with the field-level yields of the key crops. The crop production estimations for a set of key crop products were obtained by combining information from interviews with small farmers, where they were asked about the yield of their products, validated with information form the technical staff in the region and from published data. The results of crop production estimation show that crop production levels (ton/ha/year) differ significantly within key crops and across reference regions.

It also shows that in many regions studied, the share of production by small farms is higher than expected.

The total estimated production by small farms in each region was also crossed with the total consumption estimates with data from ESA in order to assess what was the share of regional consumption needs that may be satisfied by small farm production. This  analysis based on the survey data was complemented with the one done at the food system level.

Relevance for monitoring and evaluation of the CAP

Data provision: A valuable crop data set has been built that could be used in the evaluation of the CAP in these areas. The approach for developing this crop data set can be replicated in other regions/countries. Official statistics and Sentinel data can be combined to produce a wide set of data in relation to crop areas and crop production.

Remote sensing technology can provide accurate and timely information on crop types, area extent and yield estimates. Such information is useful to objectively quantify the crop production capabilities of small farms and the potential impact that CAP interventions may have on these capabilities.

Another potential application could be to use this data from small farms for assessing investment measures. Given that in some countries small farms account for a very large percentage of their farm sector, and that these farms also receive CAP funding, having data on them is very useful, especially for designing programme measures. These resources are crucial in developing regional farm systems that contribute to sustainable food production, improve farm incomes, as well as, increasing the diversity of food systems, thus contributing to their resilience.

Comparison of crop productivities: Crop type maps that contains crop area and production estimates can be used to extrapolate productivity from one region to a wider territory and compare crop production productivity patterns in neighbouring regions. There are samples in different regions covered by SALSA. However, in order to actually use the methodology in other non-neighbouring regions, for instance in other countries, one may need to reproduce the data, collect the data.

Complementarity with FADN: The main value added of the SALSA crop production estimates is that they make visible what cannot be found in official sources like FADN, which does not capture small farm statistics (due to minimum threshold). Even in agricultural censuses there is a need for a minimum size so as to be included. This is the most frequent challenge for evaluators concerning FADN representativeness. This tool allows the evaluators to have at least a very good approximation of what is missing from FADN and judge if this is important in terms of UAA, production volumes and values (if average prices are applied). Irrespective of whether the evaluation refers to small farms, this is a pertinent tool for assessing FADN’s representativeness in certain areas and, if needed, complement the FADN estimates.

Policy making and evaluation: Study countries produced maps of crop productivity and changes of small farm productivity. There was also a need for explanations of why small farms produce, i.e. the enabling environment, e.g., policy measures. For this reason, in all SALSA countries there were communities of practice, which were networks consisting of organisations, small farms and small farm businesses, NGOs, policymakers. Results were discussed within these groups and the feedback was positive.  Stakeholders consider that data could be useful for future policy making and evaluation. For instance, for selecting projects or for the set up of inerventions targeted at this type of farms, which can serve to better understand the programme/measure intervention logic.

The methodology estimating production uses, partly, sentinel earth observations and requires the crop type map tool of SALSA. Thus, the usual constraints and limitations apply. The tool's adoption requires adaptation and application of the algorithms and training to recognize the crop types of the region or the Member State.

Last modification date: 
09/12/2021