Ag-Knowledge‎ > ‎

Product Overview

 Copyright Geocledian 2018

Introduction

ag|knowledge is a REST API for mobile and web applications allowing easy access and integration of satellite remote sensing data & analytics into your agricultural application. The API provides access to field monitoring products for registered parcels (= fields or parts of land). The data for each parcel will be immediately updated as soon as new measurements are available. 

Typical image products are: 
  • Visible images (True colour images, RGB) of the parcel 
  • Live green vegetation indicator (Vitality images, NDVI)
  • Vegetation variations maps
  • A variety of specialized vegetation indices that provide information about the vegetation status (e.g. Chlorophyll, Nitrogen or vegetation water content)
  • Time series of the above listed products 
  • Full history of the last 5 years for all products
Analysis functions available are: 
  • NDVI & other vegetation index time series statistics per parcel
  • Phenology statistics on a variety of specialized parameters (e.g. crop performance, start of season, anomalous vegetation behaviour?)
  • Notification messages
  • Crop type verification
  • Parcel comparisons & benchmarking
  • Others on request 
For modelling and dedicated data analysis purposes we also offer access to the underlying index and raw sensor values. Radar data is also available on request. The functionality is constantly being updated. 

Please have a look at our demo web client to get a first impression. We have a bunch of different visualization components available for free that can help you make the most out of the data (see Tools site).

ag|knowledge is free of charge for development purposes. For operational use ag|knowledge is available on an annual subscription basis. A small fee per hectare and year will be charged. To obtain a developer account please register here or contact us directly at info@geocledian.com. Let us know your area of interest and give us an indication about the domain you want to use the service for. With a developer account you will also have full access to the documentation and you will get direct support. 

ag|knowledge focuses on agriculture, but can also be used for forest monitoring or other tasks such as environmental or urban monitoring. If you have any specific needs please let us know. 

This web service is provided to you by geo|cledian. All rights reserved ©2018 geo|cledian.


API Overview - Basic Monitoing package

We provide a REST API supporting typical GET, PUT and DELETE commands. A WMS is available on request. For accessing the API you need a user key and the base URL. Using the API is simple: register your parcel and seconds later you can retrieve the information stack available for it. This stack is updated as soon as a new sensor measurements are available. Please note that it makes most sense to register parcels with one single crop. Parcels with mixed crops can be used too, of course, but some API functions may not deliver the expected results.

To register a parcel it is required to provide: 
  • the geometry (geographic coordinates) of the parcel 
  • the crop type
  • planting (seeding) date 
  • (planned) harvesting date 
It is also recommended to provide a unique parcel and farm name. Monitoring will start 4 weeks before the planting date and automatically stop 6 weeks after harvest

The sample syntax for the registration is: 
POST https://geocledian.com/agknow/parcels

Once the parcel is registered a number of products will be available. These can be accessed by a simple get request:
GET https://geocledian.com/agknow/parcels/<parcel-id>/vitality/<raster-id> 

Product: Visible (True colour) image

Characteristics: The visible (true color) image is also called an RGB (Red-Green-Blue) image according to the spectrals bands it is composed of. It shows the latest visual image of the parcel, but also all previous images can be obtained. The product is representing the normal visible colours of the object like in a photo. Light and dark green colours usually represent vegetation. Brown colours either represent open soil or ripe crops. Light brown (beige) colours may represent fully dry ripe crops or harvested fields with stubbles still on the field. Specific colours can occur during blossom. The image is generated from  reflectance values derived from the visible spectral bands of the satellite sensor. 

Use Cases: The visible RGB image is an easy to understand product for non experts. It corresponds to the visual experience in the field. By visualising a time series the crop growth status can be roughly detected on these images. Also vegetation irregularities are already visible. Particular attention needs to be payed to brown colours. Bare soil and ripe crops may look similar. A time series of images is required to allow for distinction. 

Product: Vitality index

Characteristics: 
The vitality index is based on the NDVI. NDVI stands for Normalised Difference Vegetation Index. It is a valuable quantitative vegetation monitoring tool used as indicator for the vitality of a crop in particular for the live green vegetation. Its value also strongly depends on the percentage of area covered by leaves. To generate the Vitality image an absolute colour table is applied to the calculated NDVI index values, which are ranging from -1 to 1. Dark brown represent low NDVI values on bare soil or water (close to 0 or below), yellow colours have been assigned to medium values (low vegetation). Green colours for high values are close to 1 and represent high, vital vegetation. The image is generated from reflectance values derived from the visible and near infrared spectral bands of the satellite sensor. The NDVI can be influenced by the atmospheric conditions and soil moisture. 

Use Cases: The vitality image is also usable for non experts as it is similar to the visible (true colour) representation. It shows differences in the amount and vitality of green vegetation.

Product: Variations Map

Characteristics:
The variations product highlights the variations within the monitored parcel. It uses a colour bar, which is adjusted for each parcel individually or a set of parcels of the same crop to get the optimum visual representation of the variations. The product is currently also based on the NDVI. Border effects and spike values have been eliminated. The colour ranges from dark blue (low values) to bright red (high values).

Use Cases: The variations image is a visual tool for precision farming applications. It provides to non expert users indications about zones of conformity and non-conformity. For maximum use it requires interpretation by experts such as crop consultants.

Product: NDVI Timeseries Statistics

Characteristics: 
The NDVI statistics allow to visualise and analyse a time series of the parcel from seeding to harvest. 

Use Cases: The ndvi statistics graph visualises the development of the live green vegetation over time. Experts can interpret crop type, crop stages or crop anomalies from these graphs. It can be understood also by non experts.  

Product: NDVI (Float) tif

Characteristics: The NDVI Float tif product allows obtaining the original NDVI values for the object. These values range from -1 to 1. The image format is GEOTIFF.

Use Cases: The NDVI float tif product serves as input for modelling or dedicated data analytics. 


API Overview - Professional Monitoring package

Product: Specialized Vegetation indices

Characteristics: In addition to NDVI we have a variety of selected vegetation indices in our professional monitoring package that can be used for dedicated applications.

Use Cases: These indices allow the derivation of Chlorophyll, Nitrogen, water stress or other special products for your crops. Here we try to give a brief overview about each vegetation index and potential applications:

Index
 NameApplication
Used bands
Value range
 Example
 NDVI
 Normalized Difference Vegetation Index Vegetation vitality. NDVI correlates with the amount of leaf area (LAI) of active, healthy, green vegetation. Advantage: The most widely used vegetation index. Disadvantage: It saturates at high LAI levels and therefore shows limited variation in dense fields with high biomass. It minimizes topographic effects.

Negative values of NDVI (values approaching -1) typically correspond to open water. Values around zero (-0.1 to 0.1) generally correspond to bare soil, barren areas of rock, sand, or snow. Low, positive values represent low or sparse vegetation (approximately 0.2 to 0.4), while high values indicate dense vegetation (values approaching 1).

Native resolution: 10m
Red (B04) &
NIR (B08)
[-1; 1]    
 NDRE1
 Normalized Difference Red Edge Index (v1)The Red Edge Indices are designed to estimate chlorophyll content in the canopy. More sensitivity in vegetation with high LAI. Less sensitivity to open water. There are several different formulas for the NDRE index. We selected the most appropriate ones.

Native resolution: 20m
Vegetation Red Edge 1 &
2 (B06 & B05)
[-1; 1]      
 NDRE2
 Normalized Difference Red Edge Index (v2)The Red Edge Indices are designed to estimate chlorophyll content in the canopy. More sensitivity in vegetation with high LAI. Less sensitivity to open water. There are several different formulas for the NDRE index. We selected the most appropriate ones.

Native resolution: 20m
Vegetation Red Edge 1 &
3 (B07 & B05)
[-1; 1]     
 NDWI  Normalized Difference Water IndexSensitive to Vegetation water content & water stress. Allows drought monitoring. NDWI is less sensitive to atmospheric effects than NDVI. Attention: There are also other "NDWI" indexes with different meaning.

Native resolution: 20m
NIR (B08) &
SWIR 1 (B11)
[-1; 1]     
 SAVI  Soil Adjusted Vegetation IndexReduced soil background effects. SAVI tries to correct for bare soil areas in the field. Therefore it is interesting in sparse vegetation canopies or early growing stages.

Native resolution: 20m
Red (B04) &
NIR (B08)
[-1; 1]
     
 EVI2  Enhanced Vegetation Index 2More sensitivity in late growth stage (high LAI). It tries to correct for soil background signals and atmospheric influences.

Native resolution: 20m
Vegetation Red Edge 1 (B05) &
Narrow NIR (B8A)
[-0.735; 1.25]
      
 CI-RE  Chlorophyll Index - Red EdgeVery high Chlorophyll & Nitrogen sensitivity. Canopy Chlorophyll & Nitrogen contents can be derived for a wide range of crops (e.g. potato, soybean, maize) and grassland. This can be used to plan harvests.

Native resolution: 20m
Vegetation Red Edge 1 &
3 (B07 & B05)
[-1; ca. 9]
 NPCRI Normalized Pigment Chlorophyll Ratio IndexAssociated with chlorophyll and nitrogen
content. This index can be found in precision agriculture applications. Crops with a low Nitrogen content can have a high carotenoid to chlorophyll ratio. Using the red and blue spectral bands, NPCRI captures the information needed to quantify chlorophyll and Nitrogen.

Native resolution: 10m
 Blue (B02) & Red (B04) [-1; 1]     
 

       Min                    (Index)                           Max
  
Depending on the index the colormap visualizes different value ranges.

Product: Reflectance tif

Characteristics: The reflectance product allows to access the raw reflectance values of the sensor for all spectral bands. The image format is GEOTIFF. 

Use Cases: The Reflectance product serves as input for modelling or dedicated data analytics.

Product: Management Zones Map

Characteristics: The management zones map provides the Field Managment Zones that are needed to apply precision farming actions specifically for field zones. By analyzing multiple years of crop development the zones with similar soil characteristics are derived. Zones with higher fertility are shown in red, zones with lower in blue.

Use Cases: The Management Zones Map is an important product for field management. Even if the map itself doesn’t provide the reasoning for such differences it serves as basis for optimized soil sampling, for fertilisation for irrigation or for planning other management actions. 


Other important information

Update Frequency 

The underlying sensor data will be updated every few days mainly depending on the local weather conditions and the geographical latitude. Without clouds you typically get new images every 3-5 days. Generally it can be said the further North or South of the Equator the higher the revisit rate but the worse the cloud situation. More arid areas will be updated more frequent then humid areas. Since 2016 the update rate has more then doubled and the native ground resolution has increased 9 times (see below)

Ground Resolution 

Since 2016 the native resolution has increased by a factor of 9. One pixel measured by the best satellite sensors we use covers an area of approx. 100 m2 (10 m x 10 m) Thus for a parcel of the size of 1 ha approx. 100 measured points (pixels) are available. The minimum parcel size should currently be not below 0.6 ha. It has to be taken into account that pixels from the border of the parcel may contain significant signal components from neighbouring objects and thus are not representative. These pixels have been filtered in our products.

Geographical Areas Covered

The service works globally and is fully operational. We are setting up the areas you request within a few days if it is not already available on or data base. We are already routinely covering large areas in Europe and Africa and all of India.

Image Formats 

Supported image formats are PNG and KML for visual images and GEOTIFF for measured values. In general the imagery is geocoded to Google Mercator projection (EPSG: 3857) and can be directly used within various map frameworks such as Google Maps, Mapbox or OpenLayers. Other image formats and projections can be delivered on request. 

Attribute information will be delivered in JSON format, coordinate data as GEOJSON or WKT.


Copyright Geocledian 2018
Comments