Welcome to the 11th webinar - esa-sen2agri.org€¦ · Cross-Community EVs for GEOGLAM. Definition;...

11th Sen2-Agri Webinar, 2 July 2020

The webinar will last around 1h

The slides will be available on the Sen2-Agri website in the coming 48hrs(http://www.esa-sen2agri.org/)

Presenters: Sophie Bontemps, Philippe Malcorps from UCLouvainGEO Secretariat teamMembers of the consortium available to answer your questions:Cosmin Udroiu, Cosmin Cara, Laurentiu Nicola, Florin Tutunaru from Cs-Romania

Welcome to the 11th webinar

http://www.esa-sen2agri.org/


Webinar outline

1. Welcome

2. Introduction

3. New version 2.0.2 of the system – Update & Next steps

4. GEO Knowledge Hub: a Click & Go Solution on AWS

5. New cropland / crop type definitions: JECAM and GEOGLAM initiatives

6. Ressources and next events


Sen2-Agri: open source system to deliver 4 Sen2-Agri products along the season

EARLY AREA INDICATOR

EARLY AREA INDICATOR

Binary map identifying annually cultivated land at 10m updated every month

Crop type map at 10 m for the main regional

crops including irrigated/rainfed

discrimination

Vegetation status map at 10 m delivered every

week (NDVI, LAI, pheno index)

Monthly cloud free surface reflectance composite at 10-20 m

in line with the GEOGLAM core products


Sen2-Agri systemA system designed to run in an automated near real time (and off line) mode to deliver agric. products as satellite images are ingested

Sen2Agri Ochestrator

S2 L1C Time Series

L8 L1T Time Series

L2A Time Series

AtmosphericCorrections

(MACCS)

Temporal Synthesis Processor

Temporal Synthesis ProcessorNode #1

Biophysical Indicators Processor

Biophysical Indicators ProcessorNode #2

Cropland Mask Processor

Cropland Mask ProcessorNode #3

Crop Type Map Processor

Crop Type Map ProcessorNode #4

In-situInformation

Surface Reflectance Composite

Vegetation Status

Indicators

Dynamic Annual

Cropland Mask

Crop Type Map & Extent Area

Automated Download

Scheduled

Scheduled

Scheduled

Scheduled

Scheduled

Triggered

=> orchestratorconcept


2 operating modes

Automated mode

a) based on the Orchestrator with by-default parmeterization, automatic data download and processing until the end of the season, on-time delivery => operational scenarios

b) Processor execution on user request, with by-default parameterization

Manual mode: manual processing, independently for any processor, without installing and configuring the whole system

System can be operated through GUI, SNAP, terminal window / command line

Sen2Agri Ochestrator

S2 L1C Time Series

L8 L1T Time Series

L2A Time Series

AtmosphericCorrections

(MACCS)

Temporal Synthesis Processor

Temporal Synthesis ProcessorNode #1

Biophysical Indicators Processor

Biophysical Indicators ProcessorNode #2

Cropland Mask Processor

Cropland Mask ProcessorNode #3

Crop Type Map Processor

Crop Type Map ProcessorNode #4

In-situInformation

Surface Reflectance Composite

Vegetation Status

Indicators

Dynamic Annual

Cropland Mask

Crop Type Map & Extent Area

Automated Download

Scheduled

Scheduled

Scheduled

Scheduled

Scheduled

Triggered


ARL7/8 for the Sen2-Agri system in the NASA Application Readiness Level (ARL) Milestones

esa-sen2agri.orgfor system downloadand forum, FAQs, Newsletters, publications


Sen2-Agri community

Top 10 downloadUNITED STATES: GERMANY: ITALY: FRANCE: SOUTH AFRICA: INDIA: IRAN: SPAIN: UK: INDONESIA:

50494631252019181816


Webinar outline

1. Welcome

2. Introduction





Easy and transparent update through a dedicated update script All these evolutions in the changelog available on Sen2-Agri website

Sen2-Agri: a continuously improving system !

Version Release date Purpose1 July 2016 First beta-version release to project demo users1.6 June 2017 Pre-release of the system to project champion-users1.6.1 July 2017 First public release

1.7 November 2017New User interface functionalitiesHandling of S2B together with S2A and L8Able to use local repository of L1C data (IPT, EODC)

1.8 Apr 2018

Pre-release of the v1.8New User interface functionalitiesNew downloader (sen2agri-services application)Pre-release of a new Vegetation indices processor (LAI-FAPAR-FCOVER-NDVI)

1.8.1 May 2018 Stable release of the v1.81.8.2 July 2018 Evolutions of the downloader and of the parameters of L2A processor1.8.3 September 2018 Redirection of the downloader to the SciHub API and reconfiguration (switch from AWS)

2 May 2019 L2A processor update (MACCS to MAJA) + parallelization2.0.1 July 2019 Diverse fixed problems, among them concerning USGS API2.0.2 March 2020 Evolution to be compatible with Sen2Cor L2A products + THEIA products import


Version 2.0.2 released on the 5th of March 2020

What are the new features in Sen2-Agri version 2.0.2?• Support for using download and use of Sen2Cor L2A products

(configurable via database)• Script for importing THEIA products (MUSCATE format)• Support for FAPAR and FCOVER production after switching to INRA

algorithm in L3B processor

What has been changed/fixed in Sen2-Agri version 2.0.2?• Changed to INRA version algorithm as the default in LAI production• MAJA is now considered by default (no more MACCS)• MAJA training interval of 2 months (instead of 3)• Updated dialog for uploading in-situ data• Improved “monitoring tab”• L3B automatic scheduling from 10 days to 1 day (unprocessed L2A

products since the start of the season checked at the end of each day)

• Updates for the new changes in USGS API• Corrections for avoiding skipping one product per page in SciHub


What’s next?

• Embedding MAJA 4.2– Totally open source– Some delay in the release (expected

soon)– Will be embedded in Sen2-Agri

system

• MAJA 3.2.2 TM– Was temporary removed from the

CNES website (end of April)– 7th of May: accessible again but it

happened to be a different 3.2.2 TM version

– A few days later: update on theirwebsite with the same version as before


Webinar outline

1. Welcome

2. Introduction





The GEO Knowledge HubTowards reproducible knowledge: example with

Sen2-Agri

Florian Franziskakis Group on Earth Observations Secretariat

Organizing the knowledge created by GEO

Digital library of Earth Observation applications

Curated and linked resources into “Knowledge Packages”

Results-oriented strategy for GEO

Digital library design

● Based on InvenioRDM

● Turn-key design

● Flexible, scalable and customizable

● Open-source, stable solution

● Vendor independent

● Supported by many institutions

Example with Sen2-AgriOpen source software repositorySatellite imageryPublication

In-situ dataDocumentation

Results

Cloud platform

Challenges towards reproducibility

Agriculture Ministries

Space Agencies

Research Institutes

International Organizations

How to install ?

Run on the cloud ?

Data format ?

Sen2-Agri

Sen2-Agri “Click & Go” on AWS

Thank you for your attention

Florian Franziskakis

GEO Secretariat

[email protected] - earthobservations.org


Webinar outline

1. Welcome

2. Introduction





2020 JECAM Annual Meeting - 23 June 2020

- JECAM Guidelines: Definition of the Minimum Earth Observation Dataset Requirements

- JECAM Guidelines for cropland and crop type definition and field datacollection

- FAO Classification of crops

⇒ GREAT SUCCESS IN TERMS OF USE AND REFERENCE IN THE LITTERATURE

⇒ “Annual Cropland” definition used as an international standard with reference to JECAM

⇒ NEED FOR UPDATE AS EO HAS CHANGED COMPLETELY: EO TIME SERIES ON CONTINUOUS BASIS (Sentinels, Planet, ….) MACHINE LEARNING CLOUD COMPUTING EXPANDS OPERATIONAL CAPACITY TO LARGE SCALE

3 JECAM guidelines available at jecam.org/documents/endorsed by the network since 2014

http://www.jecam.org/


EAVs for GEOGLAM

GEOGLAM EAV WorkshopLouvain-la-Neuve, Belgium | 23-24 October 2019

Development of requirements for a set of GEOGLAM “Essential Agriculture Variables (EAVs)” to leverage and complement other community efforts to define their thematic “essential variables.”

On-going process at GEOGLAM level with some working sessions planned this weekMore info

Core EAVs Supporting EAVs Cross-Community EVs for GEOGLAM

Definition Remotely-sensed, policy-relevant indicators of state and change of agricultural land use and productivity from national to global scales.

Remote sensing-derived “building blocks” that can be combined with other information or data to provide Core EAVs.

Supporting EAVs that explicitly leverage work

already done by other EV communities

http://earthobservations.org/geoglam.php?t=eo_data_coordination&s1=eodc_eav_wg


http://earthobservations.org/geoglam.php?t=eo_data_coordination&s1=eodc_eav_wg

24

Essential Agricultural Variables

Core EAVs Definition Agricultural Land Type Covered Application or Policy Supported

Utilizedagricultural areas mask

Binary determination of currently cultivated lands

all annual and perennial, including permanent (>5

years grass), nonpermanent, and managed grassland

Land cover/use state & change; Broad brush monitoring,

especially in early months, to narrow analysis, mask products

for general assessment

Annual CropMask

Binary determination of all lands with active

agricultural development

all annual crops (JECAM definition)

Land cover/use state & change; Broad brush monitoring to

narrow analysis, mask products for general assessment

Crop type map

Determination of locations currently growing crop

types, expressed as a map

Each nationally relevant crop, accounting for ~80%

of total area under production

Crop type area estimate

Determination of areal extent of currently growing crop types, expressed as a

unit of area

Production; interannual land cover/use change

Crop Yield estimation

Harvestable weight* of commodity per unit area*

(*definitions must be declared)

Each nationally relevant crop type, accounting for ~80% of total area under

production

Productivity, markets & trade, insurance, humanitarian

Crop yield forecast

Within season, pre-harvest forecast of harvestable

weight* of commoity per unit area* (*definitions

must be declared)

Each nationally relevant crop group or crop type, as appropriate, accounting for

~80% of total area under production

Productivity, markets & trade, insurance, humanitarian

Cropcondition

assessment

Measure (quantitative? qualitative?) of crop status

relative to short-term reference* (*definition

must be declared)

For crop group or crop type, as each season progresses

Productivity, markets & trade, insurance, humanitarian, pest

and disease detection

Cover cropDetermination of cultivation

of a crop between main cropping seasons, not for the purpose of harvesting

Crop rotation

The sequence of crop groups or crop type cultivated for the purpose of harvesting, within or

between years

can not be a separate EAV but a ag. practice as a cover crop can be also harvested

Towards an Updated JECAM Guidelines for cropland and crop type definitionsubmitted for interactive discussion and possible JECAM endorsement (June 2020 ?)




JECAM for Essential Agricultural Variables

Agriculture Mask

Crop Mask Rangeland MaskMulti-Year Fallow Mask

Annual CropMask

Perennial CropMask

Field Delineation

Crop Type Map

Crop Management (attributes) Irrigation / rainfed …Tillage / no tillage …Catch / cover … cropResidues mngt (burning

…)Annual fallow

Crop productionCrop type area estimate

Yield forecastYield estimate

Annual CropType Map

Perennial CropType Map

Crop condition



JECAM Guidelines for cropland and crop type definition and field data collection (2014)The annual cropland from a remote sensing perspective is a piece of land of minimum 0.25 ha (min. width of 30 m) that is sowed/planted and harvestable at least once within the 12 months after the sowing/planting date. The annual cropland produces an herbaceous cover and is sometimes combined with some tree or woody vegetation.* **

*the herbaceous vegetation expressed as fcover (fraction of soil background covered by the living vegetation) is expectedto reach at least 30 % while the tree or woody (height >2m) cover should typically not exceed a fcover of 15%.

**There are 3 known exceptions to this definition. The first concerns the sugarcane plantation and cassava crop which areincluded in the cropland class although they have a longer vegetation cycle and are not yearly planted. Second, takenindividually, small plots such as legumes do not meet the minimum size criteria of the cropland definition. However, whenconsidered as a continuous heterogeneous field, they should be included in the cropland. The third case is the greenhousecrops that cannot be monitored by remote sensing and are thus excluded from the definition





The annual crop mask from a remote sensing perspective is defined by all pieces of land of minimum 0.25 ha (min. width of 30 m) where at least one crop is sowed/planted and harvestable once within the 12 months after the sowing/planting date. The annual cropland produces an herbaceous cover and is sometimes combined with some tree or woody vegetation.* **

Update 2 - Spatial Scale Proposition to remove the minimum from the definition but add this info:The minimum mapping unit of the annual crop mask is 25-50 m² (0,0025-0,0050 ha) for kitchen gardens and smallholder farming systems, and 100-900 m² (0,01 to 0,09 ha) for adjacent kitchen gardens and most cropping systems (still to be discussed at JECAM level).

*the herbaceous crop vegetation expressed as fcover (fraction of soil background covered by the living vegetation) is expected to reach at least 30 % while the tree or woody (height >2m) cover should typically not exceed a fcover of 15%.

**There are 3 known exceptions to this definition: - The first concerns the sugarcane plantation and cassava

crop which are included in the cropland class although they have a longer vegetation cycle and are not yearly planted.

- Second, taken individually, small plots such as legumes do not meet the minimum size criteria of the cropland definition. However, when considered as a continuous heterogeneous field, they should be included in the cropland. Second, some agroforestry plots might exceed the 15% tree or woody cover while belonging to the annual-crop mask.

- The third case is the greenhouse crops that cannot be monitored by remote sensing and are thus excluded from the definition

Update 1 – Wording changeProposition to change Annual Cropland for Annual Crop Mask :the crops are annual while the annual cropland could be ambiguous these days.




The Annual-Crop Mask is defined by all pieces of land where at least one crop is sowed/planted and fully harvestable within the 12 months after the sowing/planting date, and refers to a given reference period (typically a growing season corresponding to the targeted crops). The annual crop produce an herbaceous cover and is sometimes combined with tree, woody vegetation or perennial crops.

Update 3 – Temporal Scale Proposition: add a temporal reference as the intense use of time series, mask of annual crop could refer to

different periods as a field cultivated for a given period can be turned into fallows for the following season“refers to a season or any given reference period corresponding the growing cycle of the targeted crops. .”

ex. Annual Crop Mask for the so called Summer Grain Season in Western Cape (SAF) Annual Mask of Annual Crops for 2018-2019 spanning from August 2018 to November 2019

The Agricultural Year Mask of Annual Crops is the cumulated areas of all the possible Annual Crop Masks over an agricultural year corresponding to 12 months or more.

Add-on concept – Agricultural YearProposition to refer to the official wording of the so called ‘Agricultural Year’ used by national statistics offices to clarify what was intuitively often referred to as the former JECAM annual cropland





The Annual-Crop Mask is defined by all pieces of land where at least one crop is sowed/planted and fully harvestable within the 12 months after the sowing/planting date, and refers to a given reference period (typically a growing season corresponding to targeted crops). The annual crop produce an herbaceous cover and is sometimes combined with tree, woody vegetation or perennial crops. * ** *** *the herbaceous vegetation expressed as fcover (fraction of soil background covered by the living vegetation) is expectedto reach at least 30 % while the tree or woody (height >2m) cover should typically not exceed a fcover of 15%.** The minimum mapping unit of the annual crop mask is 25-50 m² (0,0025-0,0050 ha) for kitchen gardens and smallholder farming systems, and 100-900 m² (0,01 to 0,09 ha) for adjacent kitchen gardens and most cropping systems (still to be discussed at JECAM level).***There are 3 known exceptions to this definition. (1) The first concerns the sugarcane plantation and cassava crop which areincluded in the cropland class although they have a longer vegetation cycle and are not yearly planted. (2) Second, some agroforestry plots might exceed the 15% tree or woody cover while belonging to the annual-crop mask. (3) The third case is the greenhouse crops that cannot be monitored by remote sensing and are thus excluded from the definition.

ex. Annual-Crop Mask for the so called Summer Grain Season in Western Cape (SAF) Agricultural Year Mask of Annual-Crops for 2018-2019 spanning from August 2018 to November 2019

The Agricultural Year Mask of Annual Crops is the cumulated mask of all the Annual-Crop Masks within a so called agricultural year (typically corresponding to 12 months but possibly more).





The Crop Type Map corresponds to the piece of land where an annual or a perennial crop is identified (see definition below) and refers to a given reference time period. In case of multiple cycle over an agricultural year, either only the primary crop or the sequence of crops is reported; in case of intercropping and mixed crops in the same field, the species association is reported as long as the least developed crop contributes to at least a quarter of the overall agricultural vegetation fCover.

Annual crop types are sowed/planted at least once a year, produces an herbaceous cover reaching at least a fCover (fraction of soil background covered by the living vegetation) of 30 % and is fully harvestable within the 12 months after the sowing/planting date; it is sometimes combined with some tree or woody vegetation and while the tree or woody (height >2m) cover should typically not exceed a fCover of 15%.

Perennial crop types are multi-annual plant (crops, forages, shrubs and trees) able to regrow without being replanted for several years (at least two years) and continue to produce grains, seeds, fruits, and biomass after a single harvest. The first harvest usually take place after one year or later.





The common hierarchical typology with common aggregation levels, as proposed by JECAM (JECAM 2014. Guidelines for field data collection) and adopted by several international initiatives, should be updated.

The proposed general legend gathers the global diversity of crop types and follows a hierarchical grouping of crops to share common aggregation levels (level 1 = land cover; level 2= crop group; level 3= crop class; level 4 = sub-class). Here below, an example for levels 1 and 2, including the proposed perennial types:Annual crops

CerealsVegetables and melonsOilseed cropsRoot / tuber crops with high starch or inulin contentBeverage and spice cropsLeguminous cropsSugar cropsOther crops

Perennial cropsFruits and berries (all excluding strawberries which are not perennial)NutsOil-bearing crops (permanent only)Spices, condiments and aromatic herbsOther perennial crops (coffee, cocoa, tea, natural rubber, hops, sisal, agave, hemp, asparagus, miscanthus, others)




http://jecam.org/wp-content/uploads/2018/10/JECAM_Guidelines_for_Field_Data_Collection_v1_0.pdf


Webinar outline

1. Welcome

2. Introduction






A comprehensive set of toolsto support our community

http://www.esa-sen2agri.org/resources/faq/

http://forum.esa-sen2agri.org/

WEBSITE FAQ

FORUM

2439 posts

http://www.esa-sen2agri.org/resources/faq/

http://forum.esa-sen2agri.org/


ESA Advanced training course Material online: Presentations, videos, poster

https://eo4society.esa.int/resources/advanced-training-course-on-land-remote-sensing-with-the-focus-on-agriculture/


https://eo4society.esa.int/resources/advanced-training-course-on-land-remote-sensing-with-the-focus-on-agriculture/



• GEO Knowledge Hub Webinar in July

• Summer break for the webinars … Next dates: – September 3rd

– November 6

• Online training according to your needs

Next events

Welcome to the 11th webinar - esa-sen2agri.org€¦ · Cross-Community EVs for GEOGLAM. Definition;...

Documents

Transcript of Welcome to the 11th webinar - esa-sen2agri.org€¦ · Cross-Community EVs for GEOGLAM. Definition;...