Root Imaging and Genotype Selection in Marshlands Restoration Project

Scott Trimble

June 22, 2021 at 10:46 am | Updated March 14, 2022 at 1:31 pm | 8 min read

In an effort to restore ailing marshlands, a group of scientists turned their focus to not only selecting appropriate native species, but also understanding specific genotypes that could hold their own against the invasive common reed. Several root investigations were conducted in a bid to find the mechanism that helped the dominant genotype succeed against invasives.

Is Patch Size Important in Marshes?

The invasive common reed has changed the community composition of the marshes in NJ Hackensack Meadowlands, and patches of Spartina are the only remaining areas with native species. These Spartina patches still manage to attract a variety of wildlife, especially threatened bird species.

Large remnant patches of Spartina patens, the native high marsh plant, seem to be resisting the invasion by the non-native common reed, Phragmites australis.

Subscribe to the CID Bio-Science Weekly article series.

By submitting this form, you are consenting to receive marketing emails from: . You can revoke your consent to receive emails at any time by using the SafeUnsubscribe® link, found at the bottom of every email. Emails are serviced by Constant Contact

The Ecology Laboratory of Rutgers University wanted to explore the factors that helped these patches of S. patens survive. The Spartina patches exist in various sizes—small and large. Ecological scientists Holzapfel and Kirby wondered if large patches were more competitive than smaller patches and whether these patches defend against the aggressive growth of Phragmites.

If the large patches were indeed more competitive, then were genetics the factor behind this phenomenon? The ecologists wanted to see if these factors could be useful in choosing plants in a project for restoring the native community composition of the marshes.

Exploring Factors That Make Large Patches More Resilient

The various aspects that the ecologists investigated were part of a larger aim to understand the factors behind the mosaic of small and large patches of S. patens in the marshes. It was important to know if the patches had formed due to fragmentation of a single large habitat, as happens often in forests, or if the marsh community always patchy in nature?

The final multidisciplinary research team—which was large and included scientists Man, Pechmann, Wu, Plank, Francios, Schnell, Burdulis, and Wadhwa—investigated the following aspects:

  1. Genotypic differences between clones taken from small and large patches and finding markers to identify them through molecular analyses.
  2. Border dynamics between S.patens and Phragmites, in the medium and long term, by using transects cutting across the two species.
  3. The performance of clonal ramets from different patches in competition with Phragmites, in the laboratory and field.
  4. The fitness of Phragmites growing at the edge of existing S.patens patches.

The study was conducted for four years, between 2007 to 2010, mainly in five sites: Harrier Meadow (HM), Lyndhurst Site (LS), River Bend (RB), Hawk Property (HP), and Fish Creek (FC).

Long Term Changes in Marsh Patches

The scientists tried to piece together the patch dynamics of marshes, using a combination of historical aerial pictures, GIS technology, and field data collection. They looked for development and changes in Spartina patches over decades. They found the entire marsh area had been reduced, but it was not possible to see if small patches existed, for example, between 1930 to 2008 at River Bend.

So, the aerial pictures alone could not tell the scientists if the patches are a result of the breaking up of a large intact marsh. They had to rely on genetic information to answer this question.

Differences in Genotype

The genetic differences in the native plants of the marshes in New Jersey were not known, so the scientists had to start from scratch. They used ISSR techniques to identify molecular markers for S.patens while they examined the genetic diversity in marshes of different sizes. They used samples from three large and three small matched patches in the sites FC, RB, and HP.

The analyses showed that there was more genetic diversity among the plants in larger patches than in populations in smaller patches. While there are many possible ways this can happen, the results are not a big surprise.

Figure 1: “Principal Coordinates Analysis (PCA) of six populations of Spartina patens sampled from Meadowlands in New Jersey-based on ISSR fragments,” Holzapfel and Kirby, 2011. (Image credits: Corpus ID: 85549497)

What was surprising was the fact that samples from patches of different sizes in the same site were not related. So, although S.patens plants of small and large patches grow close to each other, they are genetically distant, except for plants in HP; see Figure 1. 

Based on these genetic differences between small and large patches in the same site, the scientists concluded that they were not a result of the fragmentation of a single large marsh. The marsh was originally patchy and these existing patches split to create the current landscape.

Border Dynamics

The team of scientists used 13 transects, each of which ran perpendicular to the S. patensPhragmites interface, so that the border lay in the center of the transect. The density, height, and cover of all the species occurring in these transects were sampled in ten sub-plots.

Figure 2: “An interface is shown for a large S. patens patch (left) and small one (right). Cover data for two Spartina patens (left site of transect) and Phragmites (right side) transects in 2007. Note the differences between the sizes of the overlap zone (O): the integral area below the 0-curve amounts to 12 for the large patch vs. 46 for the small patch,” Holzapfel and Kirby, 2011. (Image credits: Corpus ID: 85549497)

The scientists found that there was no clear-cut border between S. patens in small patches and Phragmites. The overlap (O) was large and both species were intermingled, as seen in the small patch in the Lyndhurst site in Figure 2. Whereas, in large patches, the border is well defined and there is only a little overlap in the cover of the two species, as in Fish Creek.

Over the four years, S.patens lost ground to Phragmites in the small patches; see Figure 3. In the large patches, the S. patens population remained stable, and it was the Phragmites‘ numbers that declined.

Fig. 3: “Change of transect zones from 2007 to 2010 in large and small Spartina patens patches,” Holzapfel and Kirby, 2011. (Image credits: Corpus ID: 85549497)

Competition with Phragmites

To test how competitive the plants are from small and large patches, scientists took ramets or clonal fragments from both sized patches. Once again, they chose three sites that had both large and small patches for sampling (HP, RB, and FC). Samples from a large patch in Raritan Bay were also used.

The Field Trial

The scientists took S.patens plugs with a soil auger and transferred them to a common restoration site. When the base growth was measured at the end of the season, the ramets from small patches showed significantly more vigorous growth than those from large patches. So, without any competition, small patch genotypes were the best performers.

Then, as the second part of this trial, Phragmites australis rhizomes with shoots were transplanted into the above experimental restoration sites. Large patch S. patens severely impacted the growth of Phragmites, and their aboveground biomass was lower than when they grew next to small patch S. patens.

Figure 4: “(a)Below-ground mass of Spartina types (large-patch and small-patch)
and Phragmites when grown without a competitor. (b) Total mass of Spartina types (large-patch and small-patch) and Phragmites when grown in competition,” Holzapfel and Kirby, 2011. (Image credits: Corpus ID: 85549497).

The Laboratory Trial

The competition experiment in the pots had five treatments, along the lines of the field experiment, with ramets from four sites. Pots had

  • only small patch plants,
  • only large patch plants,
  • only Phragmites australis,
  • small patch plants and Phragmites, and
  • large patch plants and Phragmites.

Before planting, the ramets were cleaned and weighed. After growing in pots for three months, the plants were harvested and separated into aboveground and belowground parts and then weighed after oven drying.

When grown alone, in absence of competition from Phragmites, the small patch S. patens had the better belowground biomass in comparison to large patch plants. It was able to accumulate more biomass than even the Phragmites, as shown in Figure 4 (a). When grown with Phragmites, small patch plants do not fare well, due to competition, and have far lower biomass. Large patch plants were not affected by competition and have similar biomass to the common reed.

Figure 5: “Root-shoot ratio of Spartina types (large-patch and small-patch) and
Phragmites when grown without a competitor,” Holzapfel and Kirby, 2011. (Image credits: Corpus ID: 85549497).

In the absence of a competitor, small patch plants with a high root to shoot ratio invest more in roots than large patch plants, while large patch plants have a lower root to shoot ratio biomass, as seen in Figure 5. Due to competition from Phragmites, which have a very vigorous root system, the root biomass of small patch plants is reduced significantly, restricting their biomass accumulation and growth potential.

Phragmites’ Performance

Leaf conductance measurements in the field show that pioneer Phragmites shoots growing in the border, among S. patens, have higher respiration rates than shoots growing in dense groups of conspecifics. This points to the existence of intraspecific competition among the common reed plants, most likely Underground.

Studying Root Dynamics with a Minirhizotron

Given the importance of root dynamics in both Spartina and Phragmites, the scientists decided to investigate it more deeply.

The scientists used minirhizotrons for direct root observation to learn about seasonal dynamics of the competition occurring belowground. The plexiglass root-tubes were placed in pure S. patens small and large patches, in zones with S. patens and Phragmites interfaces in HP, and in the restoration site. They installed nine tubes one meter deep at an angle of 45° from the soil surface. The roots of plants grow around the tubes undisturbed.

To collect images twice every month, the scientists chose the CI-600 In-Situ Root Imager, manufactured by CID Bio-Science Inc. The root scanner has a rotating scan head attached to a long probe. To take images, the root scanner is inserted into the root-tubes. At each observation, the scan head was rotated to make two high-resolution images  of different parts of the root-tube. The images were then transferred by a USB connection to a computer and analyzed by the software WinRhizotron to estimate root length, growth, and density changes over time.

Figure 6: “Root images for Hawk Property Spartina patch from April-September,
2010,” Holzapfel and Kirby, 2011. (Image credits: Corpus ID: 85549497)

The minirhizotron root scans showed that total root length increases from April to July and then declines, suggesting a cyclic trend in the root dynamics; see Figure 6. There was, however, no difference in root trends between small and large patches or at the border where the S. patens experience competition from Phragmites.

Since root system differences could be ruled out, scientists decided to check if root exudates could be the mechanism by which S. patens in large patches control Phragmites’ growth.

They took plants from the sites FC, HP, and RB, where both small and large patches existed. The plants from both sized patches were grown as monocultures, in aerated hydroponic setups in plastic containers. The scientists used distilled water in the hydroponic systems. After fourteen days of growth, a sample of water from each of the plastic containers was analyzed in the laboratory for root exudates. There was no difference between root exudates in plants of small and large patches or between sites.

So, the mechanism controlled by genetic difference, between large and small patch plants, was not uncovered by the current experiment.

Answers Lead to More Questions

The current study provided some novel and surprising insights into the patch dynamics of marshes in the region. The difference in genotypes in small and large patches in an area was unexpected. Though the study lasted for four years, the scientists could not determine how genetics was influencing plants’ interaction. The scientists have identified three main venues to explore the marsh dynamics further. However, the main aim of finding the best genotype for marsh restoration projects was answered conclusively. The large patch type, S. patens, is favored for planting as it manages to restrict the invasive Phragmites and thrive.

Vijayalaxmi Kinhal
Science Writer, CID Bio-Science
Ph.D. Ecology and Environmental Science, B.Sc Agriculture


Holzapfel, C., & Kirby. E. (2011). Clonal diversity and resistance to invasion in remnant salt marsh patches dominated by Spartina patens. Rutgers University, Fusion Ecology Lab Final Report. FAS-N/MEADOWLANDS ENVIRONMENTAL RESEARCH INSTITUTE – FELLOWS PROGRAM 2008-2010.

Request a quote for a CID Bio-Science Product

Pricing and all related materials will be sent directly to your inbox.