Assessing the clonal expansion strategy of landscape-forming plants

Vegetated coastal landscapes are often formed by clonally expanding plants. By developing rhizomes or stolons these plants can spread laterally and distribute themselves in space. In this protocol we describe a method for deriving the clonal expanding strategy of dune grasses from still images. First we will discuss a more labour intensive method that requires the excavation and reconstruction of a clonal network. Second, we will discuss a validated automated approach based on two connecting algorithms that allows for a less time-consuming expansion of a dataset.


Introduction
Many coastal landscapes emerge through two-way interrelations between landscape-forming plants and their physical environment. Clonally expanding plants can stimulate entrapment of airborne or water-suspended particles with increasing patch size and shoot density [1][2][3][4] . While rapid landscape colonization -by placing shoots further apart -can stimulate sediment accretion over a larger area, often a minimum number of closely spaced shoots is required for reducing wind or water flow below the sedimentation threshold. To investigate whether this trade-off between landscape colonization and exploitation can lead to inter-species or within-species differences in clonal expansion strategies, we developed a method to quantify the clonal network architecture and the resulting spatial shoot organization. This protocol is designed to derive step sizes between consecutive shoots in a clonal network as a proxy for clonal expansion strategy.
Probability distributions functions fitted on the derived step size data can subsequently be used to understand the patterns of clonal plant movement. As the clonal network of species can disintegrate over time -thereby complicating the reconstruction of clonal connections -we suggest investigating the clonal expansion strategy in the earlier phase of landscape colonization.

The approach
Here, a methodology is proposed to derive the spatial coordinates of individual shoots within a clonal individual and to assess the inter-shoot distances in a clonal plant network. To that end, young clonal individuals are sampled from the edge of a biogeomorphic ecosystem (e.g. embryonic dunes, pioneer marshes or expanding seagrass meadows). Before starting sampling in an area it is recommended to scan the area well, so the plants (sufficient number of replicates N~5) you choose are representative of the area. To gain some experience in handling the plants and understanding the strength or vulnerability of the rhizomal architecture, we suggest digging out a few plants and trying to trace or reconstruct their network prior to sampling.
In this protocol we will describe two methods for describing the clonal expansion strategy of plants from still images. We refer to them as being the manual or the automated approach.
The automatedapproach requires Matlab to run a script that derives the spatial coordinates of shoots from these images. It subsequently calculates their inter-shoot distributions using two connecting algorithms (i.e. traveling salesman and nearest neighbour method). The code required for this protocol is provided as Supplementary File 1 (see Source codesection under Equipment for more information).

1)
Sampling in the field: · Notebook (paper) or tablet for drawing the rhizomal network.
2) Image analyses · Photoshop or equivalent for image manipulation.
· Image J for measuring step sizes, reorientation angles and branching degrees (manual approach).

·
The algorithms for translating images to spatial coordinates and step size distributions are provided as a Matlab script (see Source codebelow).

3) Source code
The Matlab based tool (developed and tested for Matlab R2015b) for translating spatial images to shoot coordinates and subsequently calculating their step sizes is included as Supplementary file 1.
The script for analysing the images (clonal_plant_analyses.m) is provided in a ZIP package that The included functions (red_pins.m and blue_pins.m) are needed to detect the spatial coordinates of the shoots. The functions ts_solve.m is needed to derive the step sizes using the traveling salesman based approach.
In addition, three different output files for each individual plant are provided (xy_plant [1,2].csv, distancesNN_plant [1,2].csv & distancesTS_plant [1,2]). Finally, an example of a connected clonal Procedure Manually reconstructing clonal plant networks in the field 1) First select appropriate candidate plants, take a reference picture before cutting of biomass ( Figure 1a).
2) Cut off all shoots at the base (collect plant material if interested) and place a labelled pin in each shoot ( Figure 1b).
3) Deploy the calibration frame (which holds a spatial scale for translating pixels to centimetres) as level as possible. Take a photo from an appropriate distance (position the camera parallel to measurement frame to prevent measurement inaccuracies) so all four corners are visible from the image. If you are unable to shoot the whole frame in one image, than take 2 or 4 images from the example of the latter).
4) Either draw the spatial shoot organization in the grid with the labels or use a field proof tablet or computer to assign labels to each individual shoot in the picture (Figure 1c). This method was validated for two dune grass species (Ammophila arenariaand Ammophila breviligulata) only (see associated publication).
Step 1-3 need to be performed as previously described, with the exception of the labelled pins. This method requires coloured pins that can be extinguished from still images, but they don't need to be labelled. After the image is taken, the plants need to be excavated to make sure all shoots are connected. However, the precise inter-shoot connections do not need to be noted down. Instead note down which clusters of shoots are connected to one another.
Computer methods · Repeat step 7 from the manual approach and straighten the image.

Field work
Selecting appropriate sites for studying the expansion strategy of colonizing landscape-forming plants can be challenging. As many coastal ecosystems are experiencing coastal squeeze they are declining or eroding 6-8 . Analysing aerial photos of areas of interest (for example using Google Earth) can aid in identifying appropriate places for studying the clonal expansion strategy of colonizing landscapeforming plants.
This protocol explains how to derive the Euclidian distances between emerging shoots on the surface.
Therefore it does not provide the 'real' rhizomal distance or clonal plant movement. To acquire this information you need to dig out the individual shoots and measure the rhizomes between connected shoots. Also, other proxies for clonal plant network architecture such as the number of nodes or internode distances are not measured using this method (for explanation of these methods see e.g. 9-12 ).

Computer analyses
As the Matlab scripts we provided in the Supplementary file are custom-made, there are always errors that can be encountered when applying them to new species, environments etc. We recommend reading the readme.txt and the annotations in the script and run the test file before using the script on newly acquired images. Basic knowledge of Matlab and functions is necessary to solve compiling errors. Note: the automatedapproach in which the step sizes are derived using two connecting algorithms (Nearest Neighbour and Traveling Salesman method) has so far only be validated for Ammophila breviligulata and Ammophila arenaria(see associated publication). Before they can be applied to different species in different environmental settings, be sure that the step size distribution you obtain using the manual approach agrees with the automateddata.

Time Taken
The field work is best done in pairs to increase time efficiency and for safety reasons. The sampling of the plants in the field depends on the chosen spatial dimension (size of references frame) and the number of shoots within that frame. Furthermore, the manualapproach is much more time consuming than the automated approach. Typically, mapping and reconstructing the network connections of a complete clonal individual plant with ~150 shoots would take around 3 hours.
The automatedapproach, which only requires clipping of biomass, replacing them with coloured pins and rapidly digging out the shoots to identify clonal individuals, will taken about one hour.   Supplementary Files