Estimation of crop nutritional status by UAV survey for site specific crop management

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F62156489%3A43210%2F20%3A43919517" target="_blank" >RIV/62156489:43210/20:43919517 - isvavai.cz</a>

Result on the web

<a href="https://mnet.mendelu.cz/mendelnet2020/mnet_2020_full.pdf" target="_blank" >https://mnet.mendelu.cz/mendelnet2020/mnet_2020_full.pdf</a>

DOI - Digital Object Identifier

—

Result language

angličtina

Original language name

Estimation of crop nutritional status by UAV survey for site specific crop management

Original language description

This study is focused on the evaluation of UAV multispectral imaging for the diagnosis of nutritional status of winter wheat in precision agriculture. A field experiment was conducted in 2018 on two plots with the total area of 36 ha and 2019 at one plot with area 41.38 ha in ZD Kojcice (Pelhrimov, Czech Republic). During the vegetation period the field survey was carried out in stem elongation (BBCH 31) and heading (BBCH 51), both important vegetation stages for application of nitrogen fertilizers. The plant samples were taken on 56 and 53 sampling points distributed across the high - and low yielded zones and later analyzed for nitrogen content in plant tissues and total amount of above-ground biomass (fresh, dry). Simultaneously, unmanned aerial imaging was carried out by multispectral cameras Micasense RedEdge or Parrot Sequoia and the images were processed in photogrammetric software to create seamless ortho-mosaic in individual spectral bands (G, R, RE, NIR). A set of vegetation indices (NDVI, GNDVI, NDRE, NRERI, SAVI, MSAVI, EVI, EVI2, etc.) was calculated from these data and the mean value estimated by zonal statistics from 2m buffer zone around each sampling points. The statistical evaluation by correlation and regression analysis showed significant relationship between crop parameters and vegetation indices from UAV survey, thus it can be said that traditional field monitoring could be replaced by UAV survey even at a low number of calibration points. From the set of vegetation indices, NDVI showed better correlation values to estimate the amount of plant biomass, while the most sensitive vegetation index for estimation of nitrogen content in plants was NDRE index.

Czech name

—

Czech description

—

Type

D - Article in proceedings

CEP classification

—

OECD FORD branch

40106 - Agronomy, plant breeding and plant protection; (Agricultural biotechnology to be 4.4)

Project

—

Continuities

S - Specificky vyzkum na vysokych skolach

Publication year

2020

Confidentiality

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection

MendelNet 2020: Proceedings of International PhD Students Conference

ISBN

978-80-7509-765-1

ISSN

—

e-ISSN

—

Number of pages

6

Pages from-to

26-31

Publisher name

Mendelova univerzita v Brně

Place of publication

Brno

Event location

Brno

Event date

Nov 11, 2020

Type of event by nationality

WRD - Celosvětová akce

UT code for WoS article

—