Crop Health (Diseases and Pests detection)

Download Product Sheet

Product Category

• Land use
• Land cover
• Natural Disaster
• Climate Change
• Coast Management 

• Marine
• Earth's Surface Motion

Financial Domains

• Investment management 
• Risk Analysis 
• Insurance management 
• Green finance

User requirements 

UN30: Need for monitoring with accurate measurements of the growth and health of trees.

UN37: Projection of risk to portfolio assets into the future.

UN55: Detecting crop damage at the level of individual farms/fields.

Description

Crop health monitoring with EO involves analysing the distinctive energy absorption and reflection patterns of healthy and stressed plants. Stressed plants exhibit reduced energy reflection in the near-infrared (NIR) spectrum compared to healthy plants. This enables NIR to identify stressed plants, often before visible signs of stress are noticeable to farmers. Additionally, various vegetation indices like NDVI, NDWI, EVI, LAI, and FAPAR, derived from the relative reflectance of visible, NIR, and SWIR light, can be utilized to monitor stressed vegetation. These indices capture altered reflectance patterns caused by factors such as water stress, nutrient deficiencies, pests, diseases, or extreme temperatures. However, multispectral sensors used for these indicators lack the ability to differentiate specific disease types due to limited spectral discrimination. The use of hyperspectral sensors overcomes this limitation but introduces challenges such as spectral complexity and interpretation, data volume and processing requirements, coarse spatial resolution, and data pre-processing challenges.

Spatial Coverage Target

Individual farm level

Data Throughput

Rapid tasking 

Data availability

• High
• High

• Low
• Low

PRODUCT SPECIFICATIONS

Main processing steps

The crop health monitoring process begins with the acquisition of optical satellite imagery that includes near-infrared (NIR) bands, and sometimes short-wave infrared (SWIR) bands, as well as SAR imagery to provide consistent monitoring, regardless of weather conditions. Both optical and SAR imagery should be with appropriate spatial resolution for the target monitoring area. Subsequently, the process entails computing various vegetation indices, utilizing thresholding or classification methods to classify health conditions, examining temporal changes to identify variations, and finally visualizing the outcomes through thematic maps.

Input data sources

Optical: Sentinel-2, VHR based on the availability like Pleiades 1A/1B & NEO, WorldView2&3, and SPOT6/7

Radar: Sentinel-1

Supporting data: N.A

Accessibility

Sentinel-1&2:  freely and publicly available from ESA.

Optical VHR imagery: commercially available on demand from EO service providers.

Spatial resolution

Sentinel-2: 10 m

Optical VHR: ≤ 1 m

Sentinel-1: 20 m

Frequency (Temporal resolution)

Sentinel-2: 6 days

Optical VHR: Sub-daily to Daily

Sentinel-1: 6 days

Latency

< 1 Day

Geographical scale coverage

Globally

Delivery/ output format

Data type: Raster

File format: GeoTIFF

Accuracies

Thematic accuracy: 80-90%

Spatial accuracy: 1.5-2 pixels of input data

Constraints and limitations

n Cloud presence

n Challenges in fields with mixed land cover (multiple crops, bare soil, vegetation).

n Lacking the ability to differentiate specific disease types due to limited spectral discrimination.

User's level of knowledge and skills to extract information and perform further analysis on the EO products.

Skills: Essential

Knowledge: Essential

P18: Crop health (diseases and pests detection)

Download the gap analysis product sheet 

Maturity score

Mean: 2.6

STD: 0.49

Constraints and limitations

·  Cloud presence

·  Challenges in fields with mixed land cover (multiple crops, bare soil, vegetation)

·  Depending on the crop/plant/disease, the accuracy can be very low, but sufficient for some use cases

Relevant user needs

UN30: Need for monitoring with accurate measurements of the growth and health of trees.

UN37: Projection of risk to portfolio assets into the future.

UN55: Detecting crop damage at the level of individual farms/fields.

R&D gaps

·  Lacking the ability to differentiate specific disease types due to limited spectral discrimination.

·  Temporal coverage of the data from existing sensors at a high enough spatial resolution.

·  Similar spectral characteristics between pest damage and other vegetation stress factors require additional in-situ data.

·  Limitations in predictive analytics

·  When it comes to vegetation diseases, the biggest limitation in setting up an EO service is the lack of field data to validate it.

·  Lack of matureness of EO needs from stakeholders. Not clear to them what can be demanded or expected.

·  Inertia in using traditionally established analysis products, which mostly require human supervision. Greater credibility to human reports than to automatic remote monitoring.

Potential improvements drivers

·  Increased efforts in downscaling current sensor data to provide the necessary temporal coverage.

·  Additional in-situ data to calculate/validate the product in each region where it is needed.

·  Capacity building: workshops, meetings, more information about what EO can provide.

·  Improvements in models for predictive analytics.

·  Hyperspectral sensors to differentiate between different types of diseases.

Utilisation level review

Utilisation score

Mean: 2.20

STD: 0.75

No utilisation

Low utilisation

·  Unawareness of the existence of commercial EO products with better specifications

·  Lack of knowledge of executives and low-risk tolerance.

Medium utilisation

·  Unawareness of the existence of the best available commercial EO product with better specifications

High utilisation

Critical gaps related to relevant user needs

Guideline gap

UN30: Need for monitoring with accurate measurement of the growth and health of trees.

UN37: Projection of risk to portfolio assets into the future.

Utilisation gap

UN55: Detecting crop damage at the level of individual farms/fields