Grain Data Solutions (GDS), headquartered in Ontario, Canada, is an AI and satellite consulting company that aims to bridge the gap between the management of natural and farming lands and cutting-edge satellite imagery and data science technologies. Serving as the eyes from the sky, we specialize in creating dashboards, interfaces, and reports for our clients, enabling them to navigate historical and real-time ground activities. We recognize that not every business possesses the resources to hire skilled GIS and remote sensing engineers with expertise in data science. Therefore, leveraging over 25 years of combined experience, our team strives to democratize access to and understanding of satellite imagery for both non-profit organizations and industries across the globe.
From automating routine tasks to advanced analytics, AI is a game-changer in agriculture. SmthOS facilitates the effective use of these AI innovations, making modern farming more efficient and productive. #AgInnovation #AI #smthOS
2:50 yeah but that accuracy is estimated from the local stations that you trained. AI people dont understand spatial representativeness and global complexity. Empirical method like AI can't handle chaos complexity but thet still mass-produce papers because top journals recklessly support them ... quality drops. High impact factor journals dont play a role of making advancement any longer... If looking at their experimemtal set up it's insanely simple. AI developers underestimate complexity but this entire society is dragged by them because of money. Poor thing... AI cant be resolution for food or climate change inequity.
John, note there are two NDWIs, one detects water bodies, and one detect water in vegetation. We used the latter one. It's based on the paper: "NDWI A Normalized Difference Water Index for Remote Sensing of Vegetation Liquid Water From Space", by B. Gao. Let us know your thoughts on this.
@@graindatasolutions the spatial structure of all 3 index maps appears to be the same even though the colors used are different. The Gao paper you cited indicates that NDWI values increase with leaf area, and so do both of the other indices you used. NIR/Green and NDWI are less prone to saturate than NDVI but it's clear all 3 are mapping variation in leaf area. Drought and water stress are not the only factors that can cause a decrease of values for NDWI or any other plant vigor index. Therefore NDWI (or NDVI, etc) must be used with other data about rainfall or soil moisture (texture, compaction, etc) in order to *attribute* variation in index values to water deficiency. I'm not sure that NDWI represents "water content on a farm", so I asked for clarification on why you interpreted NDWI the way you did.
for such small scale i.e. at farm level did you use sentinel-2 data? How good was predicted value with actual value of crop yield while using these data ?
S2, L8 and L7, all could work well on scale bigger than a few acres. Commercial satellites are good, but we know many farmers around the world cannot afford it.
