To take on the challenge, we intend to build a near-real-time crop production monitoring and forecasting system. This can be utilised to strengthen the ability of societies in better responding to food production shocks, food price spikes, and food supply chain disruptions triggered by COVID-19. This system can disseminate information for different levels of users derived from different sources of remote sensing data. For governors and large institutes, national to global scale crop production information is vital in order to make informed decisions. While for local authorities and farmers, finer and detailed information for small farmland are required, so we have used Sentinel-2 data to provide spatially and spectrally more advanced information related to crop growth status.
The COVID-19 pandemic is set to almost double acute hunger by the end of 2020, as new World Food Programme (WFP) figures indicate an additional 130 million lives and livelihoods will be at risk.
Now of concern globally is the risk that farmers will not have access to inputs and labor for the next planting season. Due to the restrictions on movement, disruptions in trade, contractions of the virus; there have been shortages in labor, effecting the food supply chain and impacting producers, processors, traders and trucking/logistics companies in food supply chains, as well as communities globally.
Lack of availability for products such as fertilisers and pesticides, coupled with labor shortages will result in intensive farming and potential catastrophic losses in the major crop districts..
Therefore, the demand for the monitoring and forecasting of COVID-19 agricultural aftershocks is increasing worldwide, as well as the need to provide early warnings of climate variations on agricultural production.
We take advantage of the Google Earth Engine (GEE) and use its remote sensing data to monitor crops; locally and globally.
The main steps are:
Cloud platform: Google Earth Engine, Colab, GitHub
Computing language: Python, JavaScript
Achievements:
Thanks to ESA and NASA's open accessed remote sensing data and Google's cloud platforms, our system can provide the following crop monitoring results.
For detailed information, please reference the project website: https://sentinellingcrop.co Currently, people can easily access the monitoring results on GEE through internet-connected devices.
References:
[1] COVID-19 and the risk to food supply chains: How to respond?
http://www.fao.org/family-farming/detail/en/c/1268820/
[2] Risk of hunger pandemic as coronavirus set to almost double acute hunger by the end of 2020
[3] Food Security and COVID-19 https://www.worldbank.org/en/topic/agriculture/brief/food-security-and-covid-19
[4] An example of visualization of current hunger over the world. https://hungermap.wfp.org/
[5] Huang, Jianxi, et al. "Assimilation of remote sensing into crop growth models: Current status and perspectives." Agricultural and Forest Meteorology 276 (2019): 107609.
[6] Huang, Jianxi, et al. "Evaluation of regional estimates of winter wheat yield by assimilating three remotely sensed reflectance datasets into the coupled WOFOST–PROSAIL model." European journal of agronomy 102 (2019): 1-13.
[7] MacBean, Natasha, et al. "Strong constraint on modeled global carbon uptake using solar-induced chlorophyll fluorescence data." Scientific Reports 8.1 (2018): 1973.
[8] Roy, David P., et al. "A general method to normalize Landsat reflectance data to nadir BRDF adjusted reflectance." Remote Sensing of Environment 176 (2016): 255-271.
[9] Gómez-Dans, José, Philip Lewis, and Mathias Disney. "Efficient emulation of radiative transfer codes using Gaussian processes and application to land surface parameter inferences." Remote Sensing 8.2 (2016): 119.
[10] Lewis, P., et al. "An earth observation land data assimilation system (EO-LDAS)." Remote Sensing of Environment 120 (2012): 219-235.
Sentinelling Crop: https://youtu.be/EF7G-LphIE0
MODIS, Sentinel, ECMWF weather prediction.
Google Earth Engine: https://earthengine.google.com/