Category Archives: Research

AgTech, Research

The Value of Hyperspectral Data

Leave a comment

Big Data Technology

in Agricultural Research

Megan_CushnahanMegan Z Cushnahan – PhD Candidate New Zealand Centre for Precision Agriculture, Institute of Agriculture and Environment, Massey University, Palmerston North, New Zealand.

 

The use of big data technology is becoming more common in agricultural research. These emerging technologies, such as hyperspectral sensing and imaging (HIS) tools proffer to replace the expensive, laborious and time-consuming data-collection methods conventionally used in agriculture.

New approaches involve non-invasive, rapid collection of high volume, versatile data at a very high spectral and spatial resolution. Experience from other industries suggests however, that refining value from big data technologies will be a key challenge for the agricultural sector.

In order to understand the challenges and opportunities created by what we call a new data economy, the author is following a precision agriculture science team tasked with developing highly advanced hyperspectral techniques for a ‘low tech’ sector.

HSI creates multi-layered, geo-referenced data early in the science process in superabundance.  This data is created at high speed in real time and does not require expensive ground sampling.  The data is extremely versatile and has the potential for many different measurements from one record.  Early observations from the study indicate that these data traits may increase the likelihood of producing ‘surplus science’, that is, science that exceeds what was judged necessary to solve the problem as defined at project launch.

The production of superabundant and highly versatile data early in the science process appears to increase the possibility of discovering new forms of valuable knowledge (methods and solutions) during the course of an investigation. However, realizing the value of these opportunities may require a departure from the classic science model.

Under data-scarcity conditions, such surplus science would be classified as undesirable ‘project creep’. In response, we propose an alternative process based on a non-linear, iterative approach that utilises heterogeneous actors to refine value from hyperspectral data.

In addition, it is proposed that for innovation in the PA sector to make the necessary rapid advances both technically and in terms of adoption, changes are needed in the way research projects are funded and structured to accommodate a new approach to science-making.

Research

What’s “Big Data” all about?

Leave a comment

Moving from “what happened” analytics to “what will happen”

James BeechJames Beech, Senior Data Scientist BNZ/Brera Consulting Group

 

 

What’s “Big Data” all about ? This presentation will look at developments in the financial services as examples of what is coming for agriculture.

Financial services are aligning data sources to create single customer views. This requires a large overhaul of traditional systems, to build platforms to handle vast amount of disparate data (Variety, Velocity, Volume, Veracity). It offers greater understanding of customer needs so services can align the right product to the right customer using predictive modelling.

Financial services biggest change is moving from “what happened” analytics to “what will happen”.  They are using new techniques to understand causes and predictions, but to do so are building teams of data scientists and data story tellers.

Geospatial data sources and government data are enabling us to get a holistic customer view. Geospatial location of the customer base gives opportunities to align to market potential. Further, the Open Data approach by local government and NZ government reflect a new view of data and information as key public assets. Government believes making data available will drive innovation through better decision making and the creation of new services, tools, and knowledge.

An example is the ANZ Truckometer. After carefully analysing the data ANZ selected key routes and applied statistical techniques to smooth out anomalies and gaps. The result is a strong correlation between traffic flows and predicting economic growth or decline as measured by GDP data from Statistics New Zealand.

Another big data example involves Cancer identification in eye images through the application of Deep learning and Machine Learning. A predictive model was trained on 80,000 labelled images and shown to predict with 87% accuracy.  These algorithm parameters allow the ability to apply to other cyclical images.

An agricultural application would be an action oriented Agricultural Dashboard.  To get results, developers must focus on the problem being solved, not the product. Other lessons show delivery in a timely and relevant manner is crucial, it must build solid ROI cases and that the user interfaces and user experience are very important.

Focus less on “all the data” or the “perfect algorithm”. Use a short agile framework to roll out product and services. Customers will provide you the feedback for next iteration.

Farmers need end to end solutions, not part of the problem solved.

Agronomy, MicroFarm, Plant Health, Research, Sensing, Soil

Onions – Plant and Crop Modelling

Leave a comment

Understanding Variation in Onions and Potential Causes

Bruce Searle, Adrian Hunt, Isabelle Sorensen, Nathan Arnold, Yong Tan, Jian Lui   Plant and Food Research

Onion growth, development, quality and yield can vary significantly within a field. This can be observed as inter-plant variability, where two plants side by side or within very close proximity vary significantly in size and maturity or quality from each other. Additionally, spatial variability in between different areas of the field has been observed. Put these two scales of variability together and there can be significant reduction in yield and profitability for growers.

It has been estimated that a modest increase of yield from 45-50t/ha associated with a 10% reduction in size variability can increase gross margins by $1700 per hectare. Add to this the fact that variability in the field results in variability in bulb maturity and therefore storage losses, minimising variability has a strong value proposition for growers.

To minimise variability we need to know how much variability is present, what causes it and when it occurs. We used soil EM maps to identify four zones across an onion field. Within each zone we recorded variability in growth and development of individual plants to better understand plant to plant variability and how this affects overall yield variability within a field.

We also monitored crop characteristics such as leaf area across a plot and light interception to understand how yield accumulated across the different zones. Soil moisture and temperature was logged at different depths for the duration of growth.

Agronomy, AgTech, Drainage, MicroFarm, Nutrients, Plant Health, Planting, Precision Agriculture, Projects, Research, Sensing, Soil, Water

Profit Mapping Variability in Onions

Leave a comment

Profit Bands Across A Paddock

 Justin Pishief

Justin Pishief and Dan Bloomer
Centre for Land and Water

 

As part of the Onions NZ project “Benchmarking Variability in Onion Crops” a process was developed to generate yield and profit maps. This presentation explains the process using the example of a 7.3 ha paddock in Hawke’s Bay.

Data from a satellite image captured in late November were used to identify high, medium and low biomass zones.  Paddock yield samples were taken from these zones at harvest and used to generate a paddock yield map. The average yield of the paddock was estimated at 95 t/ha, with a predicted total field harvest of 669 tonnes. This compares to the grower recorded harvest of 614 tonnes.

The relative yield data were combined with grower supplied costs and returns to determine gross margins across the paddock. Data were mapped in ArcGIS and a Gross Margin map with five “profit bands” produced. The highest band had a mean Gross Margin of $11,884/ha compared to the lowest at $3,225/ha.

The breakeven gross margin yield is estimated to be 62.5 t/ha at current costs and prices. The estimated cost to business of low performing areas is $27,945, assuming the whole paddock could achieve the top band mean yield.

The poorest performing areas were identified by the grower as impacted by a failed council drain and areas of slowed drainage in the main paddock areas. An OptiSurface® assessment using historic HBRC LiDAR elevation data analysed of the impact of ponding on the site and also suggested ponding was a significant issue.

An OptiSurface® landform assessment was conducted using both single plain and optimised surface designs and the soil movement required to allow effective surface drainage was determined.

The assessment showed ponding could be avoided by land shaping with 224 m3/ha soil movement and few areas requiring more than 100 mm cut or fill. The cost is estimated at $2,000/ha or approximately $14,000 total.

Agronomy, AgTech, MicroFarm, Plant Health, Planting, Precision Agriculture, Projects, Research, Sensing

Enhancing Value of New Zealand Onions

Leave a comment

Onions New Zealand Research project

 

Dr Jane Adams
Research and Innovation Manager, Onions New Zealand Inc.

The New Zealand onion industry expects to further develop high value export markets, particularly in Asia, which could see its exports double to $200million by 2025. To realise these export opportunities the industry needs to improve efficiency and consistency of production and reliably supply high quality onions.

Currently industry average yields for brown onions vary between 33 and 50t/ha depending on season, which are significantly below demonstrated potential average yields of 100t/ha. Competition for productive land mean growers must maximise both productivity and crop value, while also meeting requirements to sustainably use resources and minimise environment impacts.

To help the industry achieve these objectives Onions New Zealand developed a project ‘Enhancing the profitability and value of NZ onions’, in collaboration with LandWISE Inc and Plant and Food Research, to understand causes of low yields and variable quality of onion crops and to develop tools to help growers monitor and manage crops. The project received additional funding from Ministry of Primary Industries Sustainable Farming Fund and commenced in July 2015.

In the first season of the project a crop of cv Rhinestone onions was grown on the LandWISE MicroFarm to allow easy access for both LandWISE and Plant and Food Research scientists to assess crop development and test methods and tools for monitoring the crop and environment at regular intervals.

Four monitoring zones were established across the trial paddock for detailed measurement of plant growth and crop development. Several tools and techniques were tested for obtaining digital data of site and crop attributes. 

An important part of the project is the involvement of local growers in discussion of progress results and use of monitoring tools and advice on crop management.  

Agronomy, LandWISE People, MicroFarm, Plant Health, Projects, Research, Soil, Tillage

MicroFarm Cover Crops Incorporated

Leave a comment

oatsvsmustard

Many thanks to Nicolle Contracting and True Earth Organics for getting our winter cover crops incorporated today.

incorporatecovercrops

This winter saw a repeat of last year’s split planting of Caliente Mustard and Oats to compare effects on soil, disease and plant growth. Seed was provided by True Earth Organics.

To gain benefit from the fumigant properties of the Caliente, it must be soil incorporated as soon as possible. This is why we have the two tractors closely following, one mulching the crop, the other incorporating the residues.

Mulching mustard - reasonable biomass, but some insect damage reducing leaf mass
Mulching mustard – reasonable biomass, but some insect damage reducing leaf mass
mulchingoats
Mulching before incorporating oats

Onions are to be planted in this area for a third season in succession. Our onion crop will also include a new area that has never had onions planted before. As part of our collaboration with Onions New Zealand and Plant and Food Research, we will compare the performance of crops in the different areas.

AgTech, Automation, Precision Agriculture, Research

Sprout: Accelerating New Technologies

Leave a comment

AgTech Start-ups

Stu BradburyStu Bradbury, Tom Rivett and Julian McCurd

Sprout is a start-up accelerator programme delivered by The BCC, Building Clever Companies in Palmerston North. The programme is designed to inspire, inform and support the next generation of Ag Tech start-ups.

Every year Sprout selects eight Ag Tech start-ups. Over 20 weeks the start-ups and entrepreneurs receive funding, alongside world class mentoring and training from leaders in technology, research and business growth.

Companies receive unparalleled access to the New Zealand and global farming network to validate and grow their businesses. At the end of the programme start-ups will have an opportunity to pitch to a hand-picked group of investors, corporate partners and potential customers to support the continuation of the rapid progress achieved through the Sprout programme.

Dunedin entrepreneurs Andrew Humphries and Tom Rivett created AgriTrack to help large scale crop farmers with the multitude of logistical challenges during harvest time, particularly those associated with vehicle management. Their solution enables live tracking of vehicles and is already being used in more than 30 farms in Western Australia.

Mangere Bridge duo Julian McCurdy and Peter Bennett set up Beez Thingz using technology to develop a platform for hive management so a network of kept bees could be accessed by everyone in the industry.

Plant Health, Projects, Research

Understanding Biology and Biofertilisers

Leave a comment

Smart Biology – Trial on your own Farm

MerfieldCharles Merfield

Future Farming Centre

There has been a phenomenal growth of biostimulants and biological fertilisers. A wide range of claims are made for these products and it can be hard for farmers and growers to tell fact from fiction.  Unfortunately, there is no rule of thumb or other simple way to separate good from bogus products: the only way is to do experimental trials.

The good news is that you don’t need lots of expensive research equipment for such experiments – the best place to do them is on your crops, your pasture and your stock.  This is because in many cases, the effects of the products are highly climate, soil, plant and animal specific – what works on one farm may not work on others.

The answers you get will also have the highest level of applicability to your business.  This is also why you should be wary of experimental results that weren’t done on a production system similar to your own property – they may be meaningless to your property.

Conducting your own experiments is also not that difficult.  There are some key things to get right including: treatments; a null control; the right duration; make sure it’s randomised and replicated, and measure what matters.

While there is a bit to learn to get such experiments right, it is not rocket science, and when you do your own experiments you can also discover a lot more about your farming and growing systems which can help you run your business better plus it puts more power in your hands.

For more information, see the on-line FFC Bulletin article or download as a pdf.

Agronomy, AgTech, Nutrients, Plant Health, Precision Agriculture, Projects, Research

On-Farm Fertiliser Applicator Calibration

Leave a comment

Guidance for farmers – check performance of fertiliser spreading

DanBloomer200

Dan Bloomer
LandWISE

Fertiliser application calibration procedures suitable for farmers applying nutrients with their own equipment have been developed.  Guidelines and a web-based calculator (see www.fertspread.nz) support on-farm checks to ensure and demonstrate application equipment is performing to expectations.

Farmers and agronomists had noticed striping in crops, especially when spreading bout widths increased to match wide sprayer bouts. Visible striping is indicative of very significant non-uniform distribution and yield loss.

A calibration check includes assessment and correcting of both application rate (kg/ha) and uniformity (CV). Farmers indicate determining the rate is reasonably easy and commonly done. Very few report completing any form of uniformity assessment.

FertSpread calculates uniformity from data from a single pass and mathematically applies overlap using both to and fro and round and round driving patterns. Test spread-pattern checks performed to date show there is a need for wider testing by farmers. Unacceptable CVs and incorrect application rates are the norm.

Fertiliser applicator manufacturers provide guidelines to calibrate equipment and some newer machines automatically adjust to correct distribution pattern based on product properties and comparing a test catch with “factory” test data.

The efficiency of catch trays is called into question. While we believe the collection tray data is acceptable to assess evenness of application, the application rate should be determined by direct measurement of weight applied to determined area.  Weighing samples involves very small quantities so scales weighing to 0.01g are required. Satisfactory options are readily available at reasonable price.

An alternative approach uses small measuring cylinders or syringe bodies to compare applied volumes. While not able to assess alternative driving patterns, this can give a direct and very visual immediate view of performance.

The Sustainable Farming Fund “On-Farm Fertiliser Applicator Calibration” project arose from repeated requests by farmers for a quick and simple way to check performance of fertiliser spreading by themselves or contractors. It was co-funded by the Foundation for Arable Research and the Fertiliser Association.

AgTech, Precision Agriculture, Research, Sensing

Turning precision data into knowledge for vegetable systems

Leave a comment

JulieOHalloran1Julie O’Halloran

 Senior Development Horticulturist, Horticulture and Forestry Science Agri-Science Queensland, Department of Agriculture and Fisheries

Queensland vegetable growers and the Department of Agriculture and Fisheries have been collaborating to adapt precision agriculture technologies into vegetable systems for the last two years.  This work has focused on 3 key areas: assessing spatial variability, implementing variable rate technologies and yield monitoring.  Significant spatial variability has been successfully identified in Queensland vegetable systems using a range of crop sensing technologies (e.g Satellite, unmanned aerial vehicle (UAV) and tractor mounted Greenseeker®).

Ground truthing the underlying causal factors of this variability has proven critical to enable informed decision making to manage block uniformity. These ground truthing activities have focused on EM38 soil mapping to understand any inherent soil variability, mapping of cut and fill areas, crop sensing imagery, strategic soil sampling programs and monitoring pest, diseases, irrigation and drainage.

While within block biomass and yield variability can be inferred from crop sensing data, it is ideal to measure yield itself.  The measurement of yield variability is currently being trialed in carrot, sweet potato and potato production in Queensland using retrofitted load-cell based, geo-referenced yield monitors.  Primarily, this provides growers with a quantitative data set of the spatial and temporal nature of yield variances and the cost of lost yield potential.

Additionally, it allows growers to cost benefit analyses of potential management interventions to improve under performing areas and make decisions as to whether these are likely to be cost effective.  This presentation will highlight the undertake outcomes from a range of variable rate applications and how multiple data layers can be used to manage crops to address spatial variability.