Establishing vegetation density by large scale sensors at the tree nursery Zulauf Schinznach Dorf

a.       Experiment name

Establishing vegetation density by large scale sensors at the tree nursery Zulauf Schinznach Dorf

b.     Literature reference

Tempfli et al., 2009 (Principles of remote sensing)

Goodchild,2010 (Scale in GIS: An overview)

Turner, 2014 (Sensing biodiversity)

www.seos-project.eu (Science Education through Earth observation for High Schools) 
Shank, 2008 (Using Remote Sensing to Map Vegetation Density on a Reclaimed Surface Mine)

c.        Output/Results. What aspect of the water cycle will your measurement/experiment cover?

Computing the density of the vegetation cover at the selected perimeter helps to calculate the evapotranspiration of the plants on-site. With the data of the UAV images (different bands), percentage of vegetation can be estimated by using the NDVI (vegetation: yes/no) and the tool ‘Raster Calculation’ in ArcGIS. With the percentage of vegetation density and the collected data from the WS-Team, effectively downscaling is possible.

d.       Where in the nursery will you perform your measurements/experiments? Describe the physical situation / the setting in the nursery where you will perform your measurements/experiments.

The location of the experiment is the area of the tree nursery Zulauf Schinznach Dorf. The borders of the system are the areas which are: i) part of the irrigation system, ii) stands on plastic (not on natural ground), iii) are not covered (green-houses). An on-site inspection and the flight experience of the tutors (ground control points of previous year experiments) will set the perimeter exactly. The disqualifying factors and the fact that the weighted plants of the WS-team must be on the aerial images will be take into account.

e.       What material will you require? Describe as precisely as possible what material you will require. Will there be costs?

To carry out the experiment following material is needed:

Unmanned Aerial Vehicle (UAV)

White cardboard

Commercial Photo camera  

Car inclusive driver (ZHAW)

Train tickets

The costs for the material and the car are taking by the ZHAW. The train tickets must be paid by the participants themselves.

f.         Where do you get the material from? Organize the material (you may coordinate with Luzi and Patrick)

The material is compiled by the tutor (M.Geilhausen) a week in advance. The participants of the experiment in week 23 2017 can join this preparation. The temporal agreement is made via E-Mail. The transport of the fragile parts will be done by car, the participants travel independently by train.

g.       What is the required timing of your measurement? Make a schedule that embeds your measurement/experiment in the week schedule.

The UAV-flight takes place in week 23, at Tuesday 6^th of June. Estimated time required: 4 hours.

h.       Are there any dependences between measurements/experiments that you must consider (your measurement depends on another experiment, such that the sequence must be considered)?

If the weather is not allowing to fly, it’s postponed to Wednesday 7^th of June. A second postpone date does not exist.

i.         Anticipated possible problems that might occur with your experiment. What is plan B if the measurements/experiments do not work out as planned?

The flight with the UAV is accomplished by a trained, experienced and instructed tutor of the ZHAW, as it is the following data processing. The weather acts as an unpredictable factor and due to the long data processing not much space is given to postpone the experiment. If the experiment can’t take place in week 23, the RS-team can use data from previous years.

j.         Are there measurement data from alternative sources that could cover a larger time interval then only the field work week?

Data from previous years can be used to include a temporal integration. The inclusion of this data sets is dependent of spare time in week 23.

k.        How do you integrate the results of your measurements/experiments with??

The results of the experiment are integrated into the calculations of the local water cycle by providing the percentage of the density of the vegetation.

Research question and Summary of possibilities

Blog2:

Hi again

This blog-post serves to keep you up do date concerning our project in Schinznach-Dorf. As the local tree-nursery is interested in getting an idea of the water flux, our team is exerted to answer this challenging but rather interesting economic question. By putting up a sound sampling design in these blogs, we’ll be able to collect significant dates until the end of May.

As a first step, we need a clear idea of what we are planning to find out. As Ruth Cohn once said: “If we have little time, let us use a lot of it at the beginning”. So it’s of considerable value for the success of a project not to rush in this phase.

After taking this advice into account, a possible research question could be:

-       Concerning weather conditions: Is the difference between the irrigated water in the tree-nursery and the backflow into the pond including evapotranspiration of the plants higher than 10 %? 

This question targets the loss of water in the irrigation system. The benefit of the 10 % threshold has to be proven, but it’s in my opinion of utmost importance to quantify the difference. This leads to a statement about the functionality of the artificial water flux, and additionally, to a quantification of its loss.

The hypothesis could be:

H0: The water cycle is closed. All water that is used for irrigation runs back, excluding the amount of evapotranspiration.

H1: The water cycle isn’t closed. There is a loss of water in the system.

To collect all the necessary data to finally answer the question shown above, our research team splits up into 1 large- and 1 small-scale sub group. Several possible research methods (measurements + computations) from the large-scale group (remote sensing) are listed up in the table below where the applicability is indicated as well.

Tabel: Index and Applicability for project Schinznach

Application / Index	+/-	Use	Notes	Applicability
NDVI (ρNIR-ρRed) /(ρNIR+ρRed)	Differs between vegetation and not veg.	Vegetation ground cover	Differs in plant types (Cherry laurel, Thuja)	++
NDWI [p(0.86um )- p(1.24um)]/ [p(0.86 um)+ p(1.24um)]	To specialised for the short period of the project	Water content in plant canopy	Plant Coverage should be nearly 100% -> highly influenced by soil background	+
GNDVI (NIR-Green) / (NIR+Green)	More specialised than NDVI (photosynthesis activity)	Comparable to NDVI	Necessary for upscaling water flux measurements	+++
DSM	For land cover classification (+ GNDVI)	Different heights	Estimate biomass (tree, shrub, perennial)	+++

To compute these indexes, spectral and morphological information is required. Establishing a sound flight plan for the UAV as well as a good strategy to record the corresponding ground truth is therefore of high importance. We will then bring together this scale by using GIS.

References:

-       - NDWI A Normalized Difference Water Index for Remote Sensing of Vegetation Liquid Water From Space (Bo-Cai Gao, 1996)

- Relationship Between Remotely-sensed Vegetation Indices, Canopy Attributes and Plant Physiological Processes: What Vegetation Indices Can and Cannot Tell Us About the Landscape (Edward P. Glenn, et.al.,2008)