Assessment of soil impact on pre- and post-harvest NDVI extrema by machine learning

Simple item page
dc.contributor.authorŁopatka, Artur
dc.contributor.authorKoza, Piotr
dc.contributor.authorSuszek-Łopatka, Beata
dc.contributor.authorSiebielec, Grzegorz
dc.contributor.authorJadczyszyn, Jan
dc.date.accessioned2024-12-11T12:26:10Z
dc.date.available2024-12-11T12:26:10Z
dc.date.issued2024-06-01
dc.description.abstractIt was observed that the difference in the maximum and minimum NDVI values at a time close toharvest (mxNDVI and mnNDVI, respectively), referred to as the haNDVI index (harvest amplitudeof NDVI), correlates with agricultural soil quality and the share of sowings. The NDVI becomes satu-rated when the values of the Leaf Area Index (LAI) signi ficantly exceed one so spatial variation inhaNDVI is mainly due to the minimum post-harvest NDVI (mnNDVI). To explain the variability ofmnNDVI values three hypotheses were formulated: i) impact of crop selection, ii) field size impact,and iii) impact of soil. To determine which of these hypotheses had the highest impact on the vari-ation in the mnNDVI, the developed machine learning models of this indicator were subjected toa test removing individual explanatory variables from them. Removing a variable does not causea signi ficant increase in model error if a variable does not contribute useful information to themodel. This test showed that the mnNDVI index depends almost exclusively on the crop indicatorwhich was the median of mnNDVI for crops, not directly from soil variables such as the agriculturalquality of soil or soil moisture. According to this, the hypothesis of direct impact of soil was rejected.The explanation for the observed correlation of haNDVI with soil quality is the agricultural practiceof choosing crops with low mnNDVI (cereals, rapeseed) at better soil conditions and crops with highmnNDVI (fodder crops, grassland) for worse soil conditions.
dc.description.sponsorshipThe project was funded by statutory activities IUNG-PIB (task 2.03)
dc.identifier.citationSoil Science Annual 75(2), 189540
dc.identifier.doi10.37501/soilsa/189540
dc.identifier.issn2300-4967
dc.identifier.urihttps://bc.iung.pl/handle/123456789/2461
dc.language.isoen
dc.publisherPolskie Towarzystwo Gleboznawcze
dc.relation.ispartofseries75; 2
dc.subjectSoil Agricultural Map
dc.subjectRemote Sensing
dc.subjectNDVI
dc.subjectCrop harvest
dc.subjectMachine Learning
dc.subjectSentinel-2
dc.titleAssessment of soil impact on pre- and post-harvest NDVI extrema by machine learning
dc.title.alternativeOcena wpływu gleby na ekstrema NDVI przed i po zbiorach za pomocą uczenia maszynowego
dc.typeArticle
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
2024 SSA haNDVI.pdf
Size:
6.22 MB
Format:
Adobe Portable Document Format
Download
Collections
Artykuły