Less Can Give You More: A Sneak-peek Into Potato Farming

About the project

A potato farmer with variable rate application machinery needed accurate data and maps of his farm to maximize crop density and health, as well as to control weed infestation. Agremo’s array of software analyses and the potato calendar confirmed his choice to analyze the remote sensed aerial images.

The farmer used a John Deere variable-rate drive planter to sow potatoes at 10 and 12 inches of intra-row spacing in the same field.

Traditional scouting methods were not useful for getting crop counts, as borders of the two plots were indistinguishable in the field. Exporting a border map from the My John Deere Operations Center to Agremo’s Stand Count Analysis provided a precise count of plants and comparison with expected populations of 18,449 plants/acre and 15,374 plants/acre for plots with 10-inches and 12-inches spacing, respectively.

To use the analysis, the farmer first obtained images of the entire field from a drone. Next, he stitched them to get a 2D map that he uploaded to the Agremo web app.

Following the same steps for all other analyses, the farmer used the Canopy Cover Analysis in the mid-season to find out which crop density showed a better growth rate. With a 96% canopy cover, the 12-inches spacing showed better plant health than the 10-inches plot that had only 89% canopy cover.

The Weed Analysis performed before the third herbicide treatment demonstrated that only 16.39% or 1.49 acres were infested. Furthermore, the SMS software allowed the farmer to prepare a prescriptive map depending on infestation intensity. He loaded the map to his John Deere tractor to enable variable herbicide spraying of his fields at the exact location of weed infestation.

Agremo’s superior artificial intelligence-driven analytics, clear guidance on the website, associated services (eg. drone operations), and help in preparing maps was a combination that aided the farmer in cutting costs, improving production, and estimating yield effortlessly.

Customer requirements and challenges

An agricultural graduate and potato farmer, whom we will call Michael Moore, decided to utilize the environmental variability within his 9.08-acres potato farm by relying on a combination of precision farming techniques.

Healthy potato crops are characterized by a large canopy with more leaves to perform photosynthesis. They also accumulate more dry matter to produce higher tuber yields.

The health of potatoes is determined by factors such as seed/tuber quality, availability of adequate inputs, the density of plants, or the occurrence of weeds. Weeds are rampant in fields, as the potatoes need high levels of nutrients but use fertilizers slowly. Moreover, less density of plants will encourage weed growth, and then compete with the crop for nutrients, light, and water, and introduce pests and diseases.

The occurrence of weeds from planting to flowering can decrease yields by 8%, and in the late harvest season by 6%.

Hence, to maximize yield, the farmer decided to optimize three agricultural operations:

Client needs

  • Crop Density: Finding the optimum crop density could prevent weed growth and maximize yield. The farmer used a spacing of 34 inches between rows, and a John Deere planter to sow potatoes with variable rates of 10 and 12 inches of intra-row spacing, to find which was best for his fields.
  • Crop Health: Monitor the potato growth and health of plants in the midseason, through estimation of the plant canopy or leaf area.
  • Weed control: Discovering infested areas early to eliminate competition with weeds.
  • To maximize yield

Existing process / Field condition

The traditional scouting method to estimate crop density would be inaccurate because the farmer couldn’t identify the exact borders of the plots with different sowing rates by walking. Moreover, it would be impossible to count plants on more than 9 acres. It would also be impossible to calculate the leaf or canopy area without instrumentation. Mapping areas with weed infestation for variable rate applications also needed a more sophisticated data collection method.

Hence, the farmer decided to use drone technology. It provided him with precise details and data from farm analysis which he needed at various phases of his potato crop.

How Agremo approached the challenge

Michael registered at the Agremo website. He identified it as the software platform for drones imagery analysis that best suited his needs, as they had a ready crop calendar for potatoes. He chose the following analyses for his farm:

Agremo is software that uses powerful processes based on Artificial Intelligence, Machine Learning, and Computer Vision. It analyzes 2D images to identify patterns and goes beyond vegetation indices to classify these patterns and quantify them. The classification is based on the difference in the color of leaves and soil, structure and size of plants, and spectral data. This way, weeds are not mistaken for crops just because they are green in color.

Agremo is a reliable partner, as they have experience analyzing over 100 species of annual and perennial crops all over the world, and are pioneers in their industry.

Design and Plan

Collect data

Analyze data

Deliver and Apply

The process and the solution

Michael followed a simple three-step-process for each of the analyses.

Drone, Map and Agremo analytics:

To get high-resolution images, he used a drone with sensors and cameras compatible with Agremo software. Using the online stitching software provided by DroneDeploy, he got a seamless map of the entire farm by merging the images taken by the drones. Following simple guidelines provided by Agremo, he uploaded the ready map onto the Agremo web app. Then he chose the analysis he wanted, specified the crop and its stage of growth

For the stand count analysis,

Michael first exported field boundaries from the My John Deere Operations Center to demarcate the plots with the two plant spacings.
The insights showed the number of plants counted and the percentage of the difference between expected and counted plants per acre, while the map showed areas with fewer plants.

Canopy cover analysis report

divided the map into two zones, one with canopy and the other without canopy, along with data on the percentage and actual acreage under each zone. The report showed Michael that

  • Plots with 10 x 34 inches spacing had 89% under canopy zones, and 11% with no canopy
  • Plots with 12 x 34 inches spacing had 96% of canopy cover and only 4% under no canopy

Weed analysis

divided the farm into three zones:

  • Fine or no stress areas
  • Potential stress
  • Weed stress

Michael had weeds on only 16.39% of the farm and knew their exact location.

Try Agremo Analytics

Free - no credit card required

What were the decisions informed by Agremo analysis?

Michael now had all the data he needed to make his decisions without having to resort to any further analysis or work on his part. He could make decisions based on precise analyses that had only a 2% error.

  • Stand count analysis: Michael was expecting 18,449 plants per acre for plots with 10-inches spacing, and 15,374 plants per acre for 12-inches spacing. He used these results later in the season to estimate his yield by taking samples from the field, measuring them, and multiplying with the number of plants.
  • Canopy cover analysis: The wider 12-inch spacing had produced plants that were bigger and healthier so they produced more canopy than closely sown potatoes. Micheal decided to use the wider spacing to grow potatoes on his farm in the future. He was also able to verify that his irrigation and fertilizer inputs were optimal.
  • Weed analysis: By using Agremo’s SMS software, it was possible to convert the shapefile into a prescriptive smart spraying map that used variable rates of herbicides for different levels of infestation. The map was uploaded to the John Deere tractor, which can read our maps to facilitate spraying.

The map helped:

  • Optimize weed control
  • Reduce herbicides and equipment use
  • Treat only weed-infested areas
  • Save money on herbicides
  • Minimize over-application of herbicides, thereby reducing their runoff or leaching into water sources
  • Lower the destruction of native plant biodiversity and bioaccumulation in beneficial insects

Summary: Return on Investment (ROI) on precision agriculture with Agremo

Table 1: Weed Analysis

Michael was amazed at how simple precision agriculture could be. Our instructions covered not just the use of the Agremo web app but had guided him throughout the entire process.

The plant numbers from the stand count analysis helped Michael estimate yield and plan his yield related processes before harvesting potatoes.

The canopy cover analysis showed an increase of 7% plant cover and an increase in productivity due to the use of 12-inch spacing.

By relying on traditional agriculture methods, treatment for the entire farm for the first and second rounds of herbicide applications costed $24.5/acre, and $16.50/acre respectively, as shown in Table 1.

Agremo weed analysis showed that only 1.94 acres out of the total 9.08 acres had weeds. Thus, $9.79/acre or a total of $88.90 for the entire farm, could be saved, after covering drone analysis costs of $4/acre.

As Michael’s experience showed, anyone can use precision farming techniques. Prior drone operating skills are not required and the Agremo web app is easy-to-use and understand.

Agremo technology can be used on any farm, however small or big, to make agricultural operations cost-effective, environmentally-sustainable, and productive.

Analyze your field

No credit card required