[UPDATED] BUKIDNON VIEWS: Information is king in agri-industry
By Roderico R. Bioco
MALAYBALAY CITY (22 October 2021) Lack of reliable information, postharvest facilities, and direct link to the market are derailing the upliftment of our agricultural industry.
Of all the inputs that can help farmers maximize yields and income, the most critical is information. Both agronomists and growers understand that access to the right data is essential to support informed crop management and marketing decisions.
With the reliance on data increasing, Smart agriculture such as using high-frequency satellite imagery is emerging as a key source of reliable information for precision agriculture, providing the ability to
- Cover large growing areas across the country
- Deliver field-level detail to observe spatial variability
- Ensure data is up-to-date across the season
- Inform decisions at key stages of the crop production cycle
We have a fragmented production system in small lots discouraging investment in larger facilities with better economy of scale.
What initiatives or advocacies can the chambers implement to further boost and revitalize Mindanao’s agricultural and aquacultural industries?
Let me focus my answer to these questions relative to the topic which is smart agriculture, hence precision farming.
We are shifting our focus to local government units (LGUs) to implement a geographic information system (GIS)-based Farm Registry System, which is part of a program that we are developing called Local Government Information System (LGIS), a universal baseline for agriculture at the local level.
With the Mandanas-Garcia Ruling, more resources will be available to the LGUs and lesser perhaps to the Department of Agriculture. It is thus imperative to motivate LGUs to invest in more precise and efficient data gathering for agriculture. This involves digitization of cadastral maps and zoning, and overlay them on production polygons identified from satellite pictures.
This will provide the foundation to do better planning, monitoring and targeting. We need to develop capacity for LGU to conduct forecasting. DA has actually started undertaking this shift, but not to the extent of providing a universal survey. For example, our Registry System for Basic Sectors in Agriculture (RSBSA) requires geo-tagging, but only to the extent of providing GPS coordinate of the location of the farmer and farm of the farmer, but not establishing the GPS coordinates of the production polygon to establish the actual production area.
RSBSA only provides production area based on anecdotal claims of the farmers who may provide inaccurate survey reply such as claiming two hectares when in fact only 1.5 hectares based on satellite image. I am guilty myself of this error.
Our aim is to integrate all these local universes and create a constellation on a national level. This will give DA and the private sector the capacity to converge macro and micro targeting. Having a solid baseline will enable us to measure properly corn crop yield, for example, and analyze the production distribution pattern, among others. We can also link the analysis with available postharvest facilities in their vicinity and identify the gap in capacity.
We can devise also more accurate Remote Sensing system which currently is quite inaccurate due to absence of baseline ground truthing. Without good baseline, margin of error is too large for RS. Machine learning system can be incorporated into RS if we have a good baseline improving AI or game theory algorithm. Artificial is not actually as it is, it is more deliberate at the start requiring more prior information before it becomes intelligent. This is how the Deep Blue was able to beat the best grandmaster in chess.
With a proper farm registry, we can establish a more precise targeted production program and link them directly to buyers. A network of postharvest or packing houses shall be developed to process all these produce before bringing to the market. Traceability issues can be resolved faster with these systems.
The GIS-based universal baseline Farm Registry System we have in mind involves among others digitizing farm lots (GPS coordinates of the corners of production polygons) identified from satellite pictures and profiling the individual farms (ground truthing) and the farmer of the lot.
I convened a small group to develop ICT solutions for LGUs to create GIS-based universal baseline for Farm Registry System for LGUs. We hope many LGUs can adopt the platform so we can stitch the data they provide to create a bigger universe (big data). This will improve Remote Sensing systems like SARAI of UPLB. It can be used also to target specific production participants, forecast yields, traceability, among others, which I hope can be utilized by Agrifoodhub of Mr. Salacup.
Satellite imagery captures light reflected from farmers’ fields. Crops reflect specific spectral signatures at different stages of the season which serve as a baseline for crop health and can indicate anomalies in the growing cycle.
If vegetation is damaged or loses vitality, the amount of reflected light changes. Those changes can be detected and mapped, helping farmers to target scouting, localize treatment, and optimize inputs for those areas. Machine learning algorithms can be incorporated to automatically report the changes.
Our team includes Dr. Edgar Po of USTP who has PhD in Geoscience and Precision Farming from Michigan State University, Mr. Allan Cuenca, ICT director of Tekconsultgrup based in US and Singapore, and Ms. Eileen Gamo, GIS-management consultant who built GIS-based administrative system of Malaybalay and Taguig cities. We will initially conduct proof of concept with two barangays in Malaybalay. One of the barangays is in low land predominantly rice and another in upland area mainly corn and high value crops.
Much of our problem with DA is lack of reliable and timely data. Our government is slow in adopting global norm in measuring using GIS and our upcoming update on Census on Agri and Fisheries (official universal baseline) by PSA in 2022 will still be using old methodology (personal interview survey system), and in the case of crops, mainly focused on survey of planting intention as a means to measure annual crop yield. No solid data.
At the planning stage, access to crop productivity maps derived from satellite imagery collections over a multi-year period can be helpful. When mined for predictive insights, these annual maps can help growers optimize field productivity by identifying in-field patterns and localized trends.
Delineating management zones in fields is an essential first step in planning the application of inputs. Vegetation indices calculated from historical imagery can be used to create baselines for in-field variation of crop productivity. Site characteristics such as topography or soil properties affect productivity and can be identified to improve planning and practices.
With this information, agronomists can set site- specific crop production goals and recommend addressing different parts of the field in appropriate ways. If crop performance is historically lower in one management zone due to nutrient deficiency or water scarcity, planning for fertilization or watering can take that information into account.
This practice not only improves yield and output prediction, but also helps farmers avoid over- application and save costs. Management zones derived from historical productivity.
The author is the regional governor of the Philippine Chamber of Commerce and Industry – Region 10. He chairs the Bukidnon Giant Bamboo Resources Corporation. This is an updated version of his statement as a reactor during the Mindanao Business Conference in August 2021. FIRST PERSON is a section in the BukidnonNews.Net website dedicated to select statements, speeches, comments and other views on public matters. If you want to contribute to FIRST PERSON, email your piece, contact details and bio profile to email@example.com.)