Contents

(Search or Access Pages Above and Below)

Crop Agriculture

General Steps to Use Imagery in GIS for Crops and Other Vegetation Purposes

This page has steps to get you started using free imagery for crop and other vegetation assessment in GIS software, including free QGIS (page on this website; opens in new tab) software. A more in depth course is planned for the future; contact us if you have questions.

* With the combination of Landsat and Sentinel-2, it is possible to acquire imagery of the same area a few times a week (depends largely on clouds). *

1) Download imagery. You can download imagery from this website located in the Free Downloads drop-down menu or download from a Recommended Free Imagery Source (page includes links to imagery sources and information regarding how to use and download viewers). If you would like to use downloads from this website, we recommend these Landsat 8 downloads and these Sentinel-2 downloads. Start off by downloading red and NIR images (band 4 and band 5 for Landsat 8; band 4 and band 8 or 8a for Sentinel-2). Some imagery can be downloaded in surface reflectance format - all can be converted to surface reflectance with the tutorials in the Landsat & Sentinel-2 AtCo/SR Tutorials drop-down menu.

2) Open GIS software and load imagery; this is shown in Free Courses.

3) The NIR, red, and SWIR bands are the most useful for vegetation (green has also been used in many applications). NIR and red are typically used for crop condition and yield forecasting (higher NIR better condition, lower red better condition); SWIR is used for plant water content applications (the lower the SWIR reflectance the more water in the plant). The red band has more uses earlier in the season, after the canopy has closed enough but prior to a dense canopy. When the canopy becomes too dense, red reflectance will saturate low, meaning that values will be similarly low throughout the field even though distinct differences in crop (or vegetation, in general) condition exist. Unlike red (or any visible band), NIR can detect differences in a denser canopy. If using red and NIR to detect crop condition within a field, it is vital that the canopy is closed enough (so soil is not too visible) and that red reflectance has not saturated. If red reflectance has saturated, just use NIR for a green field. Corn becomes challenging to assess after tasseling, due to the obscuring effect tassels have. NIR can be used solely for soybeans and other predominantly green fields with a closed canopy well into R-stages.

NIR and red are negatively correlated, whereby healthier, more robust vegetation has higher NIR surface reflectance and red has lower surface reflectance. A way to check for this is to open red and NIR images in GIS over a field of interest (images can be classified with a grayscale where lower values are darker [this is usually done automatically by software), and look to see see if area of darker pixels corresponds to areas of lighter NIR (higher reflectance). It can be difficult to see differences in pixel amounts when viewing red band imagery in GIS because values are so low (red surface reflectance in a healthy green soybean canopy, for example, is about 3%, blue is about 2.5%, green is about 5%, while NIR can be about 60%). You can symbolize the raster layer so the entire grayscale classification is from custom values that basically just encompass the crop values to help view differences better. To do this in QGIS, right-click on the image layer and select Properties to view the Layers Properties window, then select Histogram and set custom Min Max values.

You can locate relatively higher and lower area of crop condition and potential yield (or vegetation, general) without converting imagery to surface reflectance, simply by viewing higher and lower areas of pixels values. However, if you want to calculate indices, you will need to either download imagery in surface reflectance format or convert to surface reflectance. For vegetation assessment purposes, and if NIR and red are generally negatively correlating, we recommend the Wide Dynamic Range Index, where α = .1 (Gitelson, 2004). For this index, simply multiply NIR surface reflectance by .1 so that it is more equal with red surface reflectance. The index can be written as: ([NIR x .1] - red) / ([NIR x .1] + red).

4) You will want to view satellite imagery over high-resolution imagery. QGIS has a nice feature where you can load Google or Bing background imagery (Harvard University page; opens in new tab) - you should do this. ArcGIS also offers this. If you prefer, Course 1C shows how to download high-resolution imagery that can be used for a base layer.

Start with these steps, and you will independently start to learn the utility of satellite imagery.

 

Reference

Gitelson, A.A. 2004. Wide dynamic range vegetation index for remote quantification of biophysical characteristics of vegetation. Journal of Plant Physiology 161; pp. 165–173.