Cloud-Free NDVI In Agriculture: When It Is Needed & Why
Cloud-free NDVI is a beneficial technology with many fields of application, and precision agriculture in particular. The finding resolves a major issue of satellite monitoring when clouds affect NDVI values. Optical imagery provides accurate NDVI on cloudless days, but there are many more days with clouds. As of now, data accuracy and availability are possible even with an overcast sky.
Do Clouds Affect NDVI?
Clouds represent natural obstacles for optical satellite imagery. Thus, if satellite images are retrieved with optical satellites, clouds make a huge impact on image quality and, correspondingly, satellite-based analytics.
Does NDVI need to be cloud-free then? Preferably. When using optical satellite imagery to generate NDVI values for agriculture, cloud-free weather is the major condition to get reliable data. It means that if the field is covered with clouds fully or partially, the data won’t be completely accurate and, thus, reliable. More than that, clouds are not the only trouble. Even when they are not covering a field, the nearby clouds still affect the analysis results indirectly by producing shadows that may influence NDVI values.
So, how to deal with clouds when creating NDVI? Cloudless vegetation index helps resolve the issue since it uses radar satellite imagery, which allows retrieving accurate data even despite the cloud cover.
Why Do We Need Cloud-Free NDVI?
Vegetation index without clouds is used to analyze crops when clouds tamper with optical monitoring. In particular, cloud-free NDVI values apply when optical ones:
were not available completely due to total or extensive cloud cover, or
were available only partially due to partial cloud cover.
In both cases, clouds do not allow obtaining enough information for vegetation analysis. This is where cloud-free NDVI helps. Indeed, it serves to restore the missing data for the previous periods, which enables efficient field management in the long run.
In fact, the main reason to use NDVI without clouds is to properly assess not only the current field condition but the historical background as well. The matter is that the state of things at present may not reflect the situation in general, and available optical cloud-free images, sometimes, are not enough to make a weighted decision. This is a case when additional historical NDVI data is important.
Cloud-free NDVI imagery in agriculture provides insights that may be useful not only to the actual field owners but also to farmers who are about to buy some property. For example, when a farmer wants to buy a field, it is necessary to analyze both the current state and the historical data. Without cloud-free NDVI, it’s difficult to make conclusions about field productivity rates because the results at the moment may be way better than the general trend.
Therefore, the assessment of vegetation development is more accurate in a retrospective. With cloud-free NDVI, farmland owners can always get correct images of a field, even if the sky is clouded.
What Is Cloud-Free Vegetation Index On Crop Monitoring?
Cloud-free vegetation index (CfVI) is a feature that enables users to get information on their fields in cloudy weather. It is possible with the reconstruction of the satellite imagery on cloudy days in the period in-between two cloudless images. The restored cloud-free images allow tracking NDVI fluctuations within a selected time.
How Does Cloud-Free Vegetation Index Work?
On average, Sentinel-2 revisits a certain field once in 5-10 days (3 days shortest for Europe). A revisit may happen on a cloudy day, so the images will be overcast. Cloudy images don’t provide sufficient information either for proper field monitoring or vegetation analysis. Under such circumstances, a farmer may obtain just a single cloud-free optical image for the whole month, and the rest will be clouded, which is certainly not enough. This is where the new technology of cloud-free NDVI comes onto the scene and proves useful.
To generate the cloud-free vegetation index, Crop Monitoring from EOSDA processes the AMCP2 data; this satellite revisits a certain field every 1 to 3 days, which allows restoring cloudy Sentinel-2 images.
How does it work? EOSDA specialists can restore cloudy images only between two cloud-free images. It means that cloud-free NDVI can be provided over the past periods, not in the near real-time.
How Accurate Is Cloud-Free NDVI vs. Standard Optical NDVI When The Sky Is Cloudy?
The cloud-free type is a radar vegetation index, which means that it is based on radar data, not optical one. Radar signals are strong enough to pierce clouds, which makes the difference. This is why cloudy weather is no longer an issue for radars, unlike with optical imagery. Besides, data accuracy with cloud-free NDVI imagery is rather high by itself because the root-mean-square error (RMSE) is less than 6%. Furthermore, to verify their cloud-free NDVI data, EOSDA experts apply cross-validation between MODIS, Landsat-8, and Sentinel-2.
EOS Crop Monitoring
Fields monitoring platform, leveraging high-resolution satellite images to identify and react to any changes remotely!
The benefits of cloud-free NDVI images are best illustrated with use cases.
The first case presents hidden damage of waterlogging that has been revealed with cloud-free NDVI. A farmer could not acquire actual information on his field vegetation for a long period of continuous rains. In the image of 04/26/2019, the crop state is poor. The next optical NDVI images of 05/24/2019 and 06/10/2019 show that the situation significantly improved. However, almost a month-long image gap prevented field monitoring due to heavy cloud cover and a lack of cloud-free vegetation index images.
With restored cloud-free NDVI images, the farmer could assess his field’s state within the gap and understand how the crops developed: when exactly vegetation indices dropped and rose again later. These cloudless NDVI insights are necessary to get an idea of reasonable field management in the future. In other words, with the cloud-free vegetation index, a farmer gets a retrospective of when and what happened in the field, which enables him to plan further actions and schedule field operations.
Waterlogging can also provoke fungal infestation in the areas that suffered the most. Spraying the entire field means additional cost inputs and unjustified fungicide contamination. However, the restored imagery reveals the damaged zones that are potentially prone to fungicide infection. By identifying such areas this way and treating them locally, the farmer will reduce both contamination and costs.
Also, with cloud-free NDVI, it is possible to analyze how moisture spreading correlates with the field relief and elevation. It will allow agriculturalists to decide on where and what crops to plant in the field. Therefore, it will be wise to grow moisture-loving species in lower parts and moisture-sensitive ones on elevations.
Thus, the cloud-free vegetation index allows reconstructing images within the gap, which helps monitor crop development all the time. Cloud-free technology is critical for optimal field management to maintain long-term field productivity and boost crop yields.
Frost Damage Monitoring
In the second case, cloud-free NDVI proved useful in frost damage detection. At the beginning of the season, a farmer got a series of automatic alerts about significant drops in the average NDVI. However, optical NDVI images didn’t report that the situation was that bad. Confused with the discrepancy, the farmer requested the insurance company to confirm the occurrence of an insured event. Due to heavy cloud cover, the optical NDVI revealed only the damaged areas, which resulted in inaccurate alerts.
Restored cloud-free NDVI images filled the gap between 04/26/2019 and 05/23/2019 and clearly showed the low vegetation zones. The matter is that at the beginning of the season, the damaged areas were covered with snow and suffered from frost, which led to the formation of icing in certain areas. The field is protected with a forest belt in the north, and a scout could not properly notice the issue through it, even though he inspected the field from all sides. Due to the relief peculiarities, he could not see the problem zones from the other field’s sides either, so the affected areas were not identified timely. This resulted in insufficient seedling emergence (with a nearly 20% lack due to frost).
When the insurance company sent its scouts to monitor the situation, they managed to detect these zones. The farmer had to re-sow the damaged crops in order to reduce yield losses. As the restored cloud-free images reveal, it helped normalize the vegetation state in the affected areas.
How To Get Cloud-Free NDVI Data On Crop Monitoring
The option of the cloud-free vegetation index will be available via Marketplace on Crop Monitoring as an add-on. You can also request it via our feedback form. Don’t hesitate to contact our sales team experts and describe your needs and preferences. The requested cloud-free data will be delivered in the .geotiff format or through API.
EOSDA To Launch New Software: EOS Forest Monitoring
EOS Data Analytics launches a platform for satellite monitoring of forest stands to provide stakeholders with sustainable forest management tools. Software will help track deforestation and detect fire hazards, save resources and preserve the environment.
Forest fires cause whopping losses globally every year and often start due to human-related ignition. Therefore, caution and wildfire prevention measures are important. Satellite monitoring helps notice and address the problem on time.
A Free Webinar: Smart Farming Technologies In Africa
EOS Data Analytics and AgroXchange host a free webinar to discuss how to accelerate the socio-economic well-being of smallholder farmers in Africa through precision farming technologies implementation.
Soil aeration is important for proper root growth. Different methods of soil aeration promote oxygen supply to the root zone, thus increasing crop productivity. Monitoring fields helps understand aeration necessity and prevent yield losses.