Agriculture: Difference between revisions

From Climate Change AI Wiki
Content added Content deleted
VisualWikitext
No edit summary
(→‎Background Readings: Added a paper)
 
(26 intermediate revisions by 6 users not shown)
Line 1: Line 1:
[[File:Agriculture.png|thumb|A schematic of the ways that machine learning can support carbon negative agriculture, forestry, and land use.]]
''This page is about the intersection of agriculture and machine learning in the context of climate change mitigation. For an overview of agriculture as a whole, please see the [https://en.wikipedia.org/wiki/Agriculture Wikipedia page] on this topic.''[[File:Agriculture.png|thumb|A schematic of the ways that machine learning can support carbon negative agriculture, forestry, and land use. Figure from "Tackling Climate Change with Machine Learning."<ref name=":0" />]]
Plants, microbes, and other organisms have been drawing CO2 from the atmosphere for millions of years. Most of this carbon is continually broken down and recirculated through the carbon cycle, and some is stored deep underground as coal and oil, but a large amount of carbon is sequestered in the biomass of trees, peat bogs, and soil. Our current economy encourages practices that are freeing much of this sequestered carbon through deforestation and unsustainable agriculture. On top of these effects, cattle and rice farming generate methane, a greenhouse gas far more potent than CO2 itself. Overall, land use by humans is estimated to be responsible for about a quarter of global GHG emissions<ref>{{Cite book|title=Climate Change 2014: Mitigation of Climate Change. Contribution of Working Group III to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change|url=https://www.ipcc.ch/report/ar5/wg3/|date=2014|coeditors=O. Edenhofer, R. Pichs-Madruga, Y.
Sokona, E. Farahani, S. Kadner, K. Seyboth, A. Adler, I. Baum, S. Brunner, P. Eickemeier, B. Kriemann, J.
Savolainen, S. Schlomer, C. von Stechow, T. Zwickel, J.C. Minx}}</ref>(and this may be an underestimate<ref>{{Cite journal|title=Are the impacts of land use on warming underestimated in climate policy?|url=https://iopscience.iop.org/article/10.1088/1748-9326/aa836d|coauthors=Natalie M Mahowald, Daniel S Ward, Scott C Doney, Peter G Hess, and James T Randerson|date=2017}}</ref>). In addition to this direct release of carbon through human actions, the permafrost is now melting, peat bogs are drying, and forest fires are becoming more frequent as a consequence of climate change itself – all of which release yet more carbon<ref>{{Cite journal|title=The study of Earth as an integrated system|url=https://climate.nasa.gov/nasa_science/science/#:~:text=The%20Study%20of%20Earth%20as%20an%20Integrated%20System&text=Earth%20system%20science%20is%20the,whole%2C%20including%20its%20changing%20climate.}}</ref>.


As described in the paper "Tackling Climate Change with Machine Learning"<ref name=":0">{{Cite journal|last=Rolnick|first=David|last2=Donti|first2=Priya L.|last3=Kaack|first3=Lynn H.|last4=Kochanski|first4=Kelly|last5=Lacoste|first5=Alexandre|last6=Sankaran|first6=Kris|last7=Ross|first7=Andrew Slavin|last8=Milojevic-Dupont|first8=Nikola|last9=Jaques|first9=Natasha|last10=Waldman-Brown|first10=Anna|last11=Luccioni|first11=Alexandra|date=2019-11-05|title=Tackling Climate Change with Machine Learning|url=http://arxiv.org/abs/1906.05433|journal=arXiv:1906.05433 [cs, stat]}}</ref>:<blockquote>
The large scale of this problem allows for a similar scale of positive impact. According to one estimate<ref>{{Cite book|title=Drawdown: The most comprehensive plan ever proposed to reverse global warming|author=Paul Hawken|date=2015}}</ref>, about a third of GHG emissions reductions could come from better land management and agriculture. ML can play an important role in some of these areas. Precision agriculture could reduce carbon release from the soil and improve crop yield, which in turn could reduce the need for deforestation. Satellite images make it possible to estimate the amount of carbon sequestered in a given area of land, as well as track GHG emissions from it. ML can help monitor the health of forests and peatlands, predict the risk of fire, and contribute to sustainable forestry. These areas represent highly impactful applications, in particular, of sophisticated computer vision tools, though care must be taken in some cases to avoid negative consequences via the Jevons paradox.
Agriculture is responsible for about 14% of GHG emissions<ref>{{Cite book|last=Intergovernmental Panel on Climate Change|url=http://ebooks.cambridge.org/ref/id/CBO9781107415416|title=Climate Change 2014 Mitigation of Climate Change: Working Group III Contribution to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change|date=2014|publisher=Cambridge University Press|isbn=978-1-107-41541-6|location=Cambridge|doi=10.1017/cbo9781107415416}}</ref>. This might come as a surprise, since plants take up CO<sub>2</sub> from the air. However, modern industrial agriculture involves more than just growing plants. First, the land is stripped of trees, releasing carbon sequestered there. Second, the process of tilling exposes topsoil to the air, thereby releasing carbon that had been bound in soil aggregates and disrupting organisms in the soil that contribute to sequestration. Finally, because such farming practices strip soil of nutrients, nitrogen-based fertilizers must be added back to the system. Synthesizing these fertilizers consumes massive amounts of energy, about 2% of global energy consumption<ref>{{Cite journal|last=Montoya|first=Joseph H.|last2=Tsai|first2=Charlie|last3=Vojvodic|first3=Aleksandra|last4=Nørskov|first4=Jens K.|date=2015-06-10|title=The Challenge of Electrochemical Ammonia Synthesis: A New Perspective on the Role of Nitrogen Scaling Relations|url=http://dx.doi.org/10.1002/cssc.201500322|journal=ChemSusChem|volume=8|issue=13|pages=2180–2186|doi=10.1002/cssc.201500322|issn=1864-5631}}</ref>. Moreover, while some of this nitrogen is absorbed by plants or retained in the soil, some is converted to nitrous oxide, a greenhouse gas that is about 300 times more potent than CO<sub>2</sub>.
⚫
== Data ==
Satellite imagery are often useful for monitoring land use. Some widely accessed resources include,


Such industrial agriculture approaches are ultimately based on making farmland more uniform and predictable. This allows it to be managed at scale using basic automation tools like tractors, but can be both more destructive and less productive than approaches that work with the natural heterogeneity of land and crops. Increasingly, there is demand for sophisticated tools which would allow farmers to work at scale, but adapt to what the land needs. This approach is often known as “precision agriculture.”
* [https://developers.google.com/earth-engine/datasets/catalog/landsat/ Landsat]
</blockquote>
* [https://developers.google.com/earth-engine/datasets/catalog/sentinel/ Sentinel]
== Machine Learning Application Areas ==
* [https://developers.google.com/earth-engine/datasets/catalog/ Earth Engine Data Catalog]
* [https://cds.climate.copernicus.eu/cdsapp#!/dataset/satellite-methane?tab=overview Methane detection from satellite]


*[[Precision agriculture|'''Precision agriculture''']]: Typical industrial agriculture releases CO<sub>2</sub> into the atmosphere by disrupting natural soil chemistry and biodiversity, and also requires chemicals that are emissions-intensive both to produce and to use. ML can help monitor emissions; reduce the need for chemicals by pinpointing pests, diseases, and weeds; and change agricultural paradigms by controlling physical robots.
Forestry related data have also been the focus of machine learning competitions, including
* '''[[Food Security|Food security]]''': By affecting rainfall and the timing of growing seasons, climate change poses a risk to food security. Machine learning can support information gathering around food supply chains, providing early warnings about -- and triggering preventative action around -- famines.
* [[Greenhouse Gas Emissions Detection|'''Monitoring agricultural emissions''']]: Agriculture is a major contributor to greenhouse gas emissions via methane (in particular from cattle farming) and nitrous oxide (from fertilizer), in addition to carbon dioxide from soil carbon breakdown. Machine learning can help track agricultural emissions directly, in addition to monitoring crop cover and livestock to help assess emissions potential.


⚫
== Background Readings ==
* [https://www.kaggle.com/c/planet-understanding-the-amazon-from-space Planet: Understanding the Amazon from space]
* [https://idtrees.org/ IDTReeS: Integrating Data science with Trees and Remote Sensing]


* [https://doi.org/10.1016/j.agsy.2020.103016 Machine learning for large-scale crop yield forecasting]
⚫
== Methods and Software ==
⚫
* ''[https://link.springer.com/article/10.1007/s11069-017-3106-x Characterizing agricultural drought in the Karamoja subregion of Uganda with meteorological and satellite-based indices]''
⚫
Some packages for working with remote sensing data are,
* [https://www.mdpi.com/2072-4292/11/6/676 Remote Sensing Techniques for Soil Organic Carbon Estimation: A Review]


== Online Courses and Course Materials ==
⚫
* [https://medium.com/sentinel-hub/introducing-eo-learn-ab37f2869f5c eo-learn]: A python package maintained by the European Space Agency, giving easy access to imagery from Sentinel satellites, as well as utilities for data processing
* [https://github.com/mapbox/robosat robosat]: A package mapbox.
* [https://solaris.readthedocs.io/en/latest/ solaris]: A package from CosmiQ Works (SpaceNet Challenge).


* '''CGIAR-Platform for Big Data in Agriculture:''' A series of webinars for data management and data mining related to crop improvement and food security. Website [https://bigdata.cgiar.org/webinars/ here].
⚫
== Recommended Readings ==
* '''Geocomputation with R:''' An open source book for geographic data analysis using R. Website [https://geocompr.robinlovelace.net/index.html here].


== Conferences, Journals, and Professional Organizations ==
* [https://www.nature.com/articles/s41598-019-54386-6 ''Monitoring tropical forest carbon stocks and emissions using Planet satellite data'']
⚫
* ''[https://link.springer.com/article/10.1007/s11069-017-3106-x Characterizing agricultural drought in the Karamoja subregion of Uganda with meteorological and satellite-based indices]''
* ''[https://www.nature.com/articles/s41558-020-0762-8 Remote Sensing Northern Lake Methane Ebullition]''


== Community ==
=== Journals ===


*'''Computers and Electronics in Agriculture:''' International journal covering computer hardware and software for solving problems in agriculture, agronomy and horticulture. Website [https://www.journals.elsevier.com/computers-and-electronics-in-agriculture here].
⚫
=== Journals and conferences ===
*'''Precision Agriculture:''' International Journal on Advances in Precision Agriculture. Website [https://www.springer.com/journal/11119 here].
*'''Remote Sensing in Agriculture and Vegetation''': Open-access journal focusing on remote sensing with special issues related to agricultural applications. Website [https://www.mdpi.com/journal/remotesensing/sections/RSAV here].

=== Workshops ===

* '''ICLR 2020 Workshop on''' '''Computer Vision for Agriculture:''' Exposes the progress and unsolved problems of computational agriculture to the AI research community. Website [https://www.cv4gc.org/cv4a2020/ here].
* '''CVPR 2020/2021 Workshop and Prize Challenge on Agriculture-Vision:''' Present recent progress on computer vision research for tackling impactful challenges in agriculture. Website [https://www.agriculture-vision.com/ here].

=== Professional Organizations and Conferences ===

* '''ISPA:''' International Society of Precision Agriculture. Website [https://www.ispag.org/ here].
* '''ECPA 2021:''' European Conference on Precision Agriculture. Website [https://www.ecpa2021.hu/ here].
* '''ASABE:''' American Society of Agricultural and Biological Engineers. Website [https://www.asabe.org/ here].
* '''ASABE AIM 2021:''' American Society of Agricultural and Biological Engineers Annual International Meeting. Website [https://www.asabemeetings.org/ here].
* '''GRSS-IEEE:''' The IEEE Geoscience and Remote Sensing Society. Website [https://www.grss-ieee.org/ here].
* '''IGARSS 2021:''' International Geoscience and Remote Sensing Symposium. Website [https://igarss2021.com/ here].
* '''ICGIRSA 2021:''' International Conference on GIS and Remote Sensing in Agriculture. Website [https://waset.org/gis-and-remote-sensing-in-agriculture-conference-in-june-2021-in-copenhagen here].
* '''DAGM:''' The German Association for Pattern Recognition. Website [https://www.dagm.de/the-german-association-for-pattern-recognition here].
* '''DAGM GCPR 2021:''' DAGM German Conference on Pattern Recognition. Website [https://dagm-gcpr.de/ here].
* '''SPIE:''' The International Society for Optics and Photonics. Website [https://spie.org/about-spie?SSO=1 here].
* '''SPIE conference:''' SPIE conference on Autonomous Air and Ground Sensing Systems for Agricultural Optimization and Phenotyping. Website [https://spie.org/si21/conferencedetails/autonomous-sensing-systems-agricultural?utm_id=rsi21scpw&SSO=1 here].
* '''WorldAgri-Tech Innovation Summit 2021:''' International summit for agri-business networking. Website [https://worldagritechinnovation.com/ here].

⚫
=== Groups and Labs ===

* '''Australian Centre for Field Robotics:''' Robotic institute focusing on autonomous robots that can work in outdoor environment. Website [https://www.sydney.edu.au/engineering/our-research/robotics-and-intelligent-systems/australian-centre-for-field-robotics.html here].

⚫
== Library and Tools ==
⚫
Some packages for working with remote sensing data are,

⚫
* '''eo-learn:''' A python package maintained by the European Space Agency, giving easy access to imagery from Sentinel satellites, as well as utilities for data processing. [https://github.com/sentinel-hub/eo-learn Github], [https://medium.com/sentinel-hub/introducing-eo-learn-ab37f2869f5c medium-post].
* '''RSCD:''' A MATLAB toolbox for remote sensing change detection. [https://github.com/Bobholamovic/ChangeDetectionToolbox Github].
* '''geemap:''' A Python package for interactive mapping with Google Earth Engine. [https://github.com/giswqs/geemap/tree/master/examples Github], [https://geemap.org/ website].
* A list of Python and R codes and different resources for geospatial analysis and EO data. [https://github.com/acgeospatial/awesome-earthobservation-code Github].
* An updated list of geospatial analysis tools. [https://github.com/sacridini/Awesome-Geospatial Github].

⚫
== Data ==


=== Remote sensing ===
* [https://www.grss-ieee.org/earthvision2020/ EARTHVISION] is regularly held at computer vision conferences.
[[Remote Sensing|Remote Sensing Datasets]] offers great opportunity to monitor agriculture and can be georeferenced to match ground measurements. In particular, [https://earthengine.google.com/ Google earth engine] offers a convenient interface over freely available satellite imagery such as [https://developers.google.com/earth-engine/datasets/catalog/landsat/ Landsat] and [https://developers.google.com/earth-engine/datasets/catalog/sentinel/ Sentinel].
* [https://groups.google.com/forum/#!category-topic/ml-news/b56zx6rIw2Q Space and AI] is organized by the ESA-CLAIRE AI Special Interest Group on Space.


=== Societies and organizations ===
=== Remote Crop Identification ===


* '''Agriculture-Vision: A Large Aerial Image Database for Agricultural Pattern Analysis:''' A dataset composed of 94,986 aerial images from 3,432 farmlands across the United States. The images contains RGB channels and Near-infrared at a resolution of 10 cm per pixel. [https://www.agriculture-vision.com/dataset Dataset], and [https://arxiv.org/abs/2001.01306 paper] describing the dataset.
=== Past and upcoming events ===
* LandCoverNet: A multispectral satellite imagery dataset acquired from Sentinel-2 and can be used for land cover classification. The dataset was created by the Radiant Earth Foundation. [http://registry.mlhub.earth/10.34911/rdnt.d2ce8i/ Dataset] and [https://radiant-mlhub.s3-us-west-2.amazonaws.com/landcovernet/Documentation.pdf documentation].
* '''CropHarvest''': The [https://openreview.net/forum?id=JtjzUXPEaCu CropHarvest dataset], compiled by researchers affiliated with NASA Harvest, provides a unified global dataset, API, and benchmarks for ML methods in crop-type mapping.


== Important considerations ==
=== Kaggle datasets ===


* [https://www.kaggle.com/vbookshelf/v2-plant-seedlings-dataset Plant Seedlings Dataset]
== Next steps ==
* [https://www.kaggle.com/aman2000jaiswal/agriculture-crop-images Agriculture crop images]
* [https://www.kaggle.com/unitednations/global-food-agriculture-statistics Global Food & Agriculture Statistics]
* [https://www.kaggle.com/fpeccia/weed-detection-in-soybean-crops Weed Detection in Soybean Crops]
* [https://www.kaggle.com/ravirajsinh45/crop-and-weed-detection-data-with-bounding-boxes crop and weed detection data with bounding boxes]
* [https://www.kaggle.com/coreylammie/deepweedsx DeepWeedsX]
* [https://www.kaggle.com/saraivaufc/california-crop-mapping-2014 California Crop Mapping - 2014]


== References ==
== References ==

Latest revision as of 07:39, 25 October 2023

This page is about the intersection of agriculture and machine learning in the context of climate change mitigation. For an overview of agriculture as a whole, please see the Wikipedia page on this topic.

A schematic of the ways that machine learning can support carbon negative agriculture, forestry, and land use. Figure from "Tackling Climate Change with Machine Learning."[1]

As described in the paper "Tackling Climate Change with Machine Learning"[1]:

Agriculture is responsible for about 14% of GHG emissions[2]. This might come as a surprise, since plants take up CO2 from the air. However, modern industrial agriculture involves more than just growing plants. First, the land is stripped of trees, releasing carbon sequestered there. Second, the process of tilling exposes topsoil to the air, thereby releasing carbon that had been bound in soil aggregates and disrupting organisms in the soil that contribute to sequestration. Finally, because such farming practices strip soil of nutrients, nitrogen-based fertilizers must be added back to the system. Synthesizing these fertilizers consumes massive amounts of energy, about 2% of global energy consumption[3]. Moreover, while some of this nitrogen is absorbed by plants or retained in the soil, some is converted to nitrous oxide, a greenhouse gas that is about 300 times more potent than CO2.

Such industrial agriculture approaches are ultimately based on making farmland more uniform and predictable. This allows it to be managed at scale using basic automation tools like tractors, but can be both more destructive and less productive than approaches that work with the natural heterogeneity of land and crops. Increasingly, there is demand for sophisticated tools which would allow farmers to work at scale, but adapt to what the land needs. This approach is often known as “precision agriculture.”

Machine Learning Application Areas[edit | edit source]

  • Precision agriculture: Typical industrial agriculture releases CO2 into the atmosphere by disrupting natural soil chemistry and biodiversity, and also requires chemicals that are emissions-intensive both to produce and to use. ML can help monitor emissions; reduce the need for chemicals by pinpointing pests, diseases, and weeds; and change agricultural paradigms by controlling physical robots.
  • Food security: By affecting rainfall and the timing of growing seasons, climate change poses a risk to food security. Machine learning can support information gathering around food supply chains, providing early warnings about -- and triggering preventative action around -- famines.
  • Monitoring agricultural emissions: Agriculture is a major contributor to greenhouse gas emissions via methane (in particular from cattle farming) and nitrous oxide (from fertilizer), in addition to carbon dioxide from soil carbon breakdown. Machine learning can help track agricultural emissions directly, in addition to monitoring crop cover and livestock to help assess emissions potential.

Background Readings[edit | edit source]

Online Courses and Course Materials[edit | edit source]

  • CGIAR-Platform for Big Data in Agriculture: A series of webinars for data management and data mining related to crop improvement and food security. Website here.
  • Geocomputation with R: An open source book for geographic data analysis using R. Website here.

Conferences, Journals, and Professional Organizations[edit | edit source]

Journals[edit | edit source]

  • Computers and Electronics in Agriculture: International journal covering computer hardware and software for solving problems in agriculture, agronomy and horticulture. Website here.
  • Precision Agriculture: International Journal on Advances in Precision Agriculture. Website here.
  • Remote Sensing in Agriculture and Vegetation: Open-access journal focusing on remote sensing with special issues related to agricultural applications. Website here.

Workshops[edit | edit source]

  • ICLR 2020 Workshop on Computer Vision for Agriculture: Exposes the progress and unsolved problems of computational agriculture to the AI research community. Website here.
  • CVPR 2020/2021 Workshop and Prize Challenge on Agriculture-Vision: Present recent progress on computer vision research for tackling impactful challenges in agriculture. Website here.

Professional Organizations and Conferences[edit | edit source]

  • ISPA: International Society of Precision Agriculture. Website here.
  • ECPA 2021: European Conference on Precision Agriculture. Website here.
  • ASABE: American Society of Agricultural and Biological Engineers. Website here.
  • ASABE AIM 2021: American Society of Agricultural and Biological Engineers Annual International Meeting. Website here.
  • GRSS-IEEE: The IEEE Geoscience and Remote Sensing Society. Website here.
  • IGARSS 2021: International Geoscience and Remote Sensing Symposium. Website here.
  • ICGIRSA 2021: International Conference on GIS and Remote Sensing in Agriculture. Website here.
  • DAGM: The German Association for Pattern Recognition. Website here.
  • DAGM GCPR 2021: DAGM German Conference on Pattern Recognition. Website here.
  • SPIE: The International Society for Optics and Photonics. Website here.
  • SPIE conference: SPIE conference on Autonomous Air and Ground Sensing Systems for Agricultural Optimization and Phenotyping. Website here.
  • WorldAgri-Tech Innovation Summit 2021: International summit for agri-business networking. Website here.

Groups and Labs[edit | edit source]

  • Australian Centre for Field Robotics: Robotic institute focusing on autonomous robots that can work in outdoor environment. Website here.

Library and Tools[edit | edit source]

Some packages for working with remote sensing data are,

  • eo-learn: A python package maintained by the European Space Agency, giving easy access to imagery from Sentinel satellites, as well as utilities for data processing. Github, medium-post.
  • RSCD: A MATLAB toolbox for remote sensing change detection. Github.
  • geemap: A Python package for interactive mapping with Google Earth Engine. Github, website.
  • A list of Python and R codes and different resources for geospatial analysis and EO data. Github.
  • An updated list of geospatial analysis tools. Github.

Data[edit | edit source]

Remote sensing[edit | edit source]

Remote Sensing Datasets offers great opportunity to monitor agriculture and can be georeferenced to match ground measurements. In particular, Google earth engine offers a convenient interface over freely available satellite imagery such as Landsat and Sentinel.

Remote Crop Identification[edit | edit source]

  • Agriculture-Vision: A Large Aerial Image Database for Agricultural Pattern Analysis: A dataset composed of 94,986 aerial images from 3,432 farmlands across the United States. The images contains RGB channels and Near-infrared at a resolution of 10 cm per pixel. Dataset, and paper describing the dataset.
  • LandCoverNet: A multispectral satellite imagery dataset acquired from Sentinel-2 and can be used for land cover classification. The dataset was created by the Radiant Earth Foundation. Dataset and documentation.
  • CropHarvest: The CropHarvest dataset, compiled by researchers affiliated with NASA Harvest, provides a unified global dataset, API, and benchmarks for ML methods in crop-type mapping.

Kaggle datasets[edit | edit source]

References[edit | edit source]

  1. ↑ 1.0 1.1 Rolnick, David; Donti, Priya L.; Kaack, Lynn H.; Kochanski, Kelly; Lacoste, Alexandre; Sankaran, Kris; Ross, Andrew Slavin; Milojevic-Dupont, Nikola; Jaques, Natasha; Waldman-Brown, Anna; Luccioni, Alexandra (2019-11-05). "Tackling Climate Change with Machine Learning". arXiv:1906.05433 [cs, stat].
  2. ↑ Intergovernmental Panel on Climate Change (2014). Climate Change 2014 Mitigation of Climate Change: Working Group III Contribution to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge: Cambridge University Press. doi:10.1017/cbo9781107415416. ISBN 978-1-107-41541-6.
  3. ↑ Montoya, Joseph H.; Tsai, Charlie; Vojvodic, Aleksandra; Nørskov, Jens K. (2015-06-10). "The Challenge of Electrochemical Ammonia Synthesis: A New Perspective on the Role of Nitrogen Scaling Relations". ChemSusChem. 8 (13): 2180–2186. doi:10.1002/cssc.201500322. ISSN 1864-5631.
Retrieved from "https://wiki.climatechange.ai/wiki/Agriculture?oldid=6148"