Spatiotemporal analysis of virus spread in vineyards

[MariaZumkeller]

I am looking to gain more information on the rate of spread of virus in vineyards through spatiotemporal analysis and am not sure which tools are best fit for my needs...

I have 5 years of data, and each year is comprised of points representing grapevines infected with virus. I've utilized point density tool to create a heat map of all points, but I am looking for more information on the rate of spread since I am especially interested in learning as much as possible of where we can expect the virus to move next year... any help is much appreciated!

An example of the point data from 2022 and 2021 is attached for reference.

[DanPatterson]

Do you know the virus is spread? (contact? wind? insects? etc).

Have you looked at whether adjacent vines get infected from one year to the next?

What happens to a vine that is infected in year 1? is it kept? removed?

[MariaZumkeller]

Thanks for your reply. The virus is primary spread by insect (vine mealybug specifically). It is not always the case that vines adjacent to infected ones are also positive as we have tested this. When a vine is identified as infected it is removed in the winter. Thus, each year “new” points are new infections. Was looking into the nearest neighbor tool to determine how far new infections are found compared to the previous years points but I’m not sure how to interpret the results for this context?  

[DanPatterson]

perhaps Generate Near Table (Analysis)—ArcGIS Pro | Documentation

to compare 2 years worth of data for the closest (first) would be interesting since you get the

• from - to id values
• their distance
• a vector angle

the x, y coordinates are just references to the year 1 and year 2 point coordinates.

You could then do some mapping, graphing of the distances and angles.

Now of course, this doesn't mean that plant A caused the infection in plant B in the next year... but it is a start

ADDENDUM

Coordinate Table To 2-Point Line (Defense)—ArcGIS Pro | Documentation

you could use the coordinates in the table to actually produce the movement vector so you can look at lines instead of points