Page 94 - Special Topic Session (STS) - Volume 1

STS407 Étienne Saint-Pierre
ii) Need for more horizontal and timely information
The expectations of data users are rapidly changing. Data users need
current, detailed and easily accessible data and analysis. The broad policies
and issues related to the agriculture sector go well beyond the primary
sector. The Census of agriculture needs to be adapted accordingly.
Whether it is to transport efficiently the agricultural goods from farms to
markets by getting real-time production and capacity data, to assess the
impact of climate change on current practices or to support the development
of sustainable ones, to develop new food products or export to international
markets, to measure the food availability and security and their impact on
health outcomes, stakeholders expect the statistical agency to produce more
frequently high-quality, granular agriculture data fully integrated with data
from other economic sectors to support evidence-based decision making.

iii) Proliferation of data and refinement of large datasets processing
techniques
Increased access to administrative and transactional data and access to
high-quality satellite images, supported by refined techniques to process large
datasets, make it possible to consider new ways to produce timely relevant
statistics with minimal contact with respondents.
There has been an increase in the availability of alternative sources of
agricultural data. So far more than 300 data sources are available to be used
at different stages of the survey cycle in the ASP. These data sources include
sources from the Canada Revenue Agency (e.g., tax data) and supply-managed
sectors (including dairy, chicken, eggs and turkey), where datasets include
quota and production figures.
Crop insurance data, which detail what crops have been planted and
insured, as well as their yield at the field level, represent a great source of
ground truth information that can be combined with satellite images, agro-
climatic data, and advanced modelling techniques to produce frequent high
quality yield and crop area estimates at a very granular level of geography
without having to contact farmers. Access to animal traceability database,
tracking movement and typical approaches in machine learning can be used
to create animal level daily movement data and real-time pork inventory from
the aggregate movement information. Leveraging these new techniques and
these alternatives sources of information will definitely transform the way the
Census of Agriculture was traditionally conducted.
Migration of the Census of Agriculture to the corporate platform used for
the production of Economic Statistics (IBSP) is a first essential step to adapt
the program to the rapidly evolving context.

83 | I S I W S C 2 0 1 9