# Detecting nitrogen levels; best indices? **URL:** \ **Category:** Agriculture\ **Created:** [March 27, 2023, 1:06pm UTC](https://community.pix4d.com/t/detecting-nitrogen-levels-best-indices/25897 "2023-03-27T13:06:13Z")\ **Posts on this page:** 14\ **Page:** 1
**Author:** ![lowaltitude](https://avatars.discourse-cdn.com/v4/letter/l/f19dbf/32.png) [@lowaltitude](https://community.pix4d.com/u/lowaltitude)\ **Post date:** [March 27, 2023, 1:06pm UTC](https://community.pix4d.com/t/detecting-nitrogen-levels-best-indices/25897/1 "2023-03-27T13:06:13Z")
Hi all, newbie here and learning fast. Does anyone have a suggestion on the best indices to use to detect nitrogen levels in plants, or even soil? I think this can be done using a multispectral set-up, with the crop reflectance showing high/low levels of nitrogen. But what is the best index to use? I understand this is a hot topic at the moment, with fertilizer costs on the increase. Thanks for any advice! ---
**Author:** ![Julius\_Petri](https://sea2.discourse-cdn.com/pix4d/user_avatar/community.pix4d.com/julius_petri/32/4592_2.png) [@Julius\_Petri](https://community.pix4d.com/u/Julius_Petri)\ **Post date:** [March 28, 2023, 6:55am UTC](https://community.pix4d.com/t/detecting-nitrogen-levels-best-indices/25897/2 "2023-03-28T06:55:59Z")
Early in the season you can use NDVI and later on when the canopy is closed NDRE: [![](https://global.discourse-cdn.com/pix4d/original/2X/a/a98a7c768892fd76b58eba49cb99e82e13b3b166.jpeg "Prescription maps for Variable Rate Application | PIX4Dfields Tutorials") ](https://www.youtube.com/watch?v=wWOifd3Xzo4) There also special indices like the REIP, which showed to be very sensitive to nitrogen levels. ---
**Author:** ![lowaltitude](https://avatars.discourse-cdn.com/v4/letter/l/f19dbf/32.png) [@lowaltitude](https://community.pix4d.com/u/lowaltitude)\ **Post date:** [March 28, 2023, 11:12am UTC](https://community.pix4d.com/t/detecting-nitrogen-levels-best-indices/25897/3 "2023-03-28T11:12:00Z")
Thanks. Is the REIP indice available in P4DF? ---
**Author:** ![Julius\_Petri](https://sea2.discourse-cdn.com/pix4d/user_avatar/community.pix4d.com/julius_petri/32/4592_2.png) [@Julius\_Petri](https://community.pix4d.com/u/Julius_Petri)\ **Post date:** [March 28, 2023, 12:19pm UTC](https://community.pix4d.com/t/detecting-nitrogen-levels-best-indices/25897/4 "2023-03-28T12:19:02Z")
Yes PIX4Dfields has a “custom index calculator” tool allowing you to calculate any index you like. But of course it depends if your sensor has the needed spectral bands for any given index. You can read more about the REIP index here: [IDB - Search results for »REIP«](https://www.indexdatabase.de/search/?s=REIP) ---
**Author:** ![lowaltitude](https://avatars.discourse-cdn.com/v4/letter/l/f19dbf/32.png) [@lowaltitude](https://community.pix4d.com/u/lowaltitude)\ **Post date:** [March 29, 2023, 7:16pm UTC](https://community.pix4d.com/t/detecting-nitrogen-levels-best-indices/25897/5 "2023-03-29T19:16:43Z")
Thanks. I’m not clear how to construct the formula in P4DF though. That link you kindly provided shows formula as the attached, but the Index Calculator doesn’t offer those options. The only available band is ‘Red’, no Red Edge (for example). ![Screenshot 2023-03-29 at 20.14.01](https://global.discourse-cdn.com/pix4d/original/2X/d/d7b6c75c7e02338067253c94a7c6ca1127a0beb4.png) ![Screenshot 2023-03-29 at 20.15.46](https://global.discourse-cdn.com/pix4d/original/2X/7/7c5db6af42d42870c7a7d3e0008a7a0fadc79e3a.png) How should it be transposed? It would be great if P4DF contained plenty of indices already. ---
**Author:** ![Julius\_Petri](https://sea2.discourse-cdn.com/pix4d/user_avatar/community.pix4d.com/julius_petri/32/4592_2.png) [@Julius\_Petri](https://community.pix4d.com/u/Julius_Petri)\ **Post date:** [March 29, 2023, 7:48pm UTC](https://community.pix4d.com/t/detecting-nitrogen-levels-best-indices/25897/6 "2023-03-29T19:48:16Z")
The reason is that your sensor does not have the bands required for that index. You can use the satellite Import tool, if you want to experiment with those indices, as the satellite data comes with a lot of bands. In general Pix4dfields supports the most common indices and only suggest you those which work with the bands your senor offers. ---
**Author:** ![lowaltitude](https://avatars.discourse-cdn.com/v4/letter/l/f19dbf/32.png) [@lowaltitude](https://community.pix4d.com/u/lowaltitude)\ **Post date:** [March 30, 2023, 8:05am UTC](https://community.pix4d.com/t/detecting-nitrogen-levels-best-indices/25897/7 "2023-03-30T08:05:39Z")
Understood, thank you. ---
**Author:** ![lowaltitude](https://avatars.discourse-cdn.com/v4/letter/l/f19dbf/32.png) [@lowaltitude](https://community.pix4d.com/u/lowaltitude)\ **Post date:** [May 2, 2023, 1:39pm UTC](https://community.pix4d.com/t/detecting-nitrogen-levels-best-indices/25897/8 "2023-05-02T13:39:38Z")
Here’s another one: how does P4DF deal with ‘to the power of’ in a formula? Please see attached for TSAVI: ![Screenshot 2023-05-02 at 14.36.56](https://global.discourse-cdn.com/pix4d/original/2X/d/d78c6663b6cb78d02faa2833fef07f9141b5f0aa.jpeg) What would the correct formula look like in P4DF? ---
**Author:** ![Jose.Arnaudo](https://sea2.discourse-cdn.com/pix4d/user_avatar/community.pix4d.com/jose.arnaudo/32/5966_2.png) [@Jose.Arnaudo](https://community.pix4d.com/u/Jose.Arnaudo)\ **Post date:** [May 4, 2023, 7:24am UTC](https://community.pix4d.com/t/detecting-nitrogen-levels-best-indices/25897/9 "2023-05-04T07:24:09Z")
Hi, You should type (1+ 0.33^2). Best, ---
**Author:** ![2323235](https://sea2.discourse-cdn.com/pix4d/user_avatar/community.pix4d.com/2323235/32/8679_2.png) [@2323235](https://community.pix4d.com/u/2323235)\ **Post date:** [September 2, 2024, 8:27pm UTC](https://community.pix4d.com/t/detecting-nitrogen-levels-best-indices/25897/10 "2024-09-02T20:27:00Z")
Hello! @lowaltitude ! How was your experience detecting nitrogen levels? What index wroked the best for you? ---
**Author:** ![lowaltitude](https://avatars.discourse-cdn.com/v4/letter/l/f19dbf/32.png) [@lowaltitude](https://community.pix4d.com/u/lowaltitude)\ **Post date:** [September 17, 2024, 8:14am UTC](https://community.pix4d.com/t/detecting-nitrogen-levels-best-indices/25897/11 "2024-09-17T08:14:46Z")
Hi. We use NDVI at the start of the season, NDRE as the canopy is full, and have started exploring NIRv which is meant to be more sensitive than NDVI. ---
**Author:** ![Franz](https://avatars.discourse-cdn.com/v4/letter/f/c37758/32.png) [@Franz](https://community.pix4d.com/u/Franz)\ **Post date:** [December 20, 2024, 8:54am UTC](https://community.pix4d.com/t/detecting-nitrogen-levels-best-indices/25897/12 "2024-12-20T08:54:01Z")
Hi @Julius_Petri, @Jose.Arnaudo, I am using a DJI Mavic 3M and have calculated the REIP using an approximation formula. When I click on “Preview” in the PIX4Dfields calculation window the index is displayed. However, when I apply the calculation, the index does not appear in the main window. Within the info section, the mean value and standard deviation (SD) are calculated and displayed numerically. Could this be a visualization issue, or is there a trick to solve this? I can provide you screenshots and the dataset for analysis upon request. Thx Franz ---
**Author:** ![Jose.Arnaudo](https://sea2.discourse-cdn.com/pix4d/user_avatar/community.pix4d.com/jose.arnaudo/32/5966_2.png) [@Jose.Arnaudo](https://community.pix4d.com/u/Jose.Arnaudo)\ **Post date:** [December 26, 2024, 10:42am UTC](https://community.pix4d.com/t/detecting-nitrogen-levels-best-indices/25897/13 "2024-12-26T10:42:43Z")
Hi, as you don’t have the correct bands for REIP index, the values you obtain might be corrupted for PIX4Dfields to display. What formula are you using? ---
**Author:** ![Jose.Arnaudo](https://sea2.discourse-cdn.com/pix4d/user_avatar/community.pix4d.com/jose.arnaudo/32/5966_2.png) [@Jose.Arnaudo](https://community.pix4d.com/u/Jose.Arnaudo)\ **Post date:** [December 26, 2024, 12:11pm UTC](https://community.pix4d.com/t/detecting-nitrogen-levels-best-indices/25897/15 "2024-12-26T12:11:06Z")
The formula doesn’t make sense. 780-700 should be reflectances referring to those bands nm… Anyway, I was able to estimate the index using your formula on fields with low biomass, make sure you properly set the histogram min and max, as well as turning Histogram equalization ON. ![image](https://global.discourse-cdn.com/pix4d/original/2X/1/1b98418ccc6ad92b14f28f2f115ab57f90f805d8.jpeg) On fields with saturated NDVI, it breaks.