Intercropping between Collard Greens and Radishes

Intercropping Between Radishes and Collard Greens

Bella Harthun

INTRODUCTION

This experiment's purpose was to determine whether collard green (Brassica oleracea var. viridis) seeds, radish (Raphanus sativus) seeds, or a mixture of both seeds grew the tallest and heaviest when exposed to different light intensities. Under varied factors, distinct plants can and will compete for access to resources.  This is also evident in the type of soil, as different soil types can increase root competition (Tomobe 2023). Since soil types that do not carry many nutrients, such as sand, are not healthy living environments, competition between plants can increase due to the lack of living materials. One of these required living materials is light exposure. In general, plants have higher growth levels when they are exposed to greater amounts of light, such as 18 hours instead of 6 (Qin 2023).  The studies that were done on varied plants, such as plants that require varied factors to thrive, such as different amounts of water and light or different soil compositions, relate back to this experiment since this experiment studies a variety of factors. These factors, while yielding the possibility of being confounding variables, can also yield discoveries on the complex growing patterns of plants. By measuring different plant species and light levels, this experiment is reflective of the ways that collards and radishes require different frequencies of variables to grow, but the inclusion of treatments containing a mixture of collard and radish seeds can also shed light on the impacts of plant competition, further yielding agricultural recommendations for plants that cooperate in close quarters, and plants that unfortunately do not.   Intercropping is a good way to test if plants can exist in certain environments, since only plants with similar needs would be able to survive in the similar environments.  Intercropping is also a good way to increase disease resistance within populations, which helps with genetic diversity (Pinheiro 2019).  Additionally, this experiment is justified by the general survivorship curve of plants: they hold a type 3 survivorship in the Plantae kingdom, meaning they reproduce quickly, they have short life spans, and they are usually at the bottom of the food pyramid. This experiment also holds the possibility of the application to varied species: it can be replicated in many ways. Not just with distinct species, but since plants have varied needs, it would be in the botanical world’s best interest to test all the variables that go into the flourishing germination of a plant. This scientific curiosity is what propels this study, and it exists to analyze how differing factors during the initial stages of a plant’s life can impact development. More specifically, light and competition between species. Comparing radish seeds and collard seeds when it comes to the optimal amounts of light needed is beneficial because whereas both collard greens and radishes belong to the same family (Brassicaceae), collards belong to the Brassica genus, whereas radishes belong to the Raphanus genus. The only taxonomic “step” before species, and fully classifying an organism, is the genus of an organism. Since collard greens and radishes are grouped in the same family but diverge at their respective genus,’ this is reflective of the small but significant differences in these plants’ makeup, as well as highlighting any variance in their nutritional needs to thrive.

If collard seeds, radish seeds, and a mixture of both are placed under differing light intensities, collards will grow the tallest, and radishes will grow the heaviest. The variables measured were light intensity and plant species, and each plant species was planted in a 9-inch pot, 3 treatments of 24 seeds of collards, 3 treatments of 24 seeds of radishes, and 3 treatments of 12 collards and 12 radishes. 1 trial of each plant species was placed under a 4-hour light cycle, an 8-hour light cycle, and a 24-hour light cycle. The total plant growth of each treatment and the seedling number that germinated during the week was measured, and at the end of the experiment, each treatment’s total weight was measured. Additionally, the average height of each seedling within each treatment was calculated and utilized to compare the possible benefits and drawbacks of differing levels of light exposure, as well as the possible impacts of plant competition while in an enclosed space.

MATERIALS AND METHODS

Experiments were conducted at Viterbo University in La Crosse, Wisconsin. The selected organisms were radishes and collard greens, and the objective was to determine the process of germination between these organisms. A 9-inch-tall black pot was used for the planting of the seeds; the pots were labelled with the appropriate species and light exposure level and the bottom of the pot was lined with a section of generic black landscaping cloth. The labeling was as follows: radishes were pots 2, 5, and 8; collards were 1, 4, and 7, and the mixed seeds were 3, 6, and 9.  The 4-hour sequence was posts 1-3, the 8-hour sequence was 4-6, and the 24-hour sequence was 7-9. The pots were then filled halfway (about 11.4 centimeters) with into a mixture of Organic Valley Topsoil and Micaflake Loose Fill Attic Insulation at a 3:1 ratio of topsoil to insulation, which is also considered vermiculite. The soil blend was moistened with distilled water and another layer of topsoil was added to leave about 4 centimeters of space from the top of the pot. The next layer was also hydrated, again with distilled water, then the seeds were sown in a 6 by 4 pattern in the soil. 3 treatments received 24 collard seeds, 3 treatments received 24 radish seeds, and the final 3 treatments received a mixture of 12 radish seeds and 12 collard greens. The planted seeds were then covered with about 1 centimeter of topsoil, and that layer was lightly watered with a spray bottle. Finally, the pots were placed into their appropriate light exposure stand locations on a Grower’s Supply Company seedling cart with 6 40-watt lightbulbs.

Throughout the experiment, the treatments were watered with a spray bottle of distilled water. Every week, the sprouting seedling count, and the total height of the surviving seedlings in each treatment were measured with a ruler in centimeters. At the consummation of the experiment, the seedlings within each treatment were excavated. The roots were washed and dried to avoid any extra weight from leftover soil, and the weight was measured in grams with a balance scale.

STATISTICS

R-Studio was utilized to analyze statistical differences in distinct factors. To test the statistical difference in the height of seedlings between plant species, an analysis of variance (ANOVA) test was done, and the p-values between each species were calculated using a pairwise t-test for the sum of the seedling height after weeks 1 through 5 (5 different ANOVA analyses and 5 different pairwise t-tests). This same process was repeated to test for a statistical difference between number of seedlings and plant species after weeks 1 and 5 (2 ANOVA analyses and 2 pairwise t-tests), between final plant mass and plant species after week 5 (1 ANOVA analysis and 1 pairwise t-test), and finally, between the average heights and weights of collards versus radishes after week 5 (2 different two-sample t-tests).  The ANOVA and pairwise t-test were used for the first 3 statistical analyses because the variable being tested was plant species, which is a categorical variable, not a binary variable. If the plant species variable was simply radishes versus collards, then it would be considered binary, and a t-test would be appropriate. Since this is not the case, an ANOVA test and pairwise t-test yield the statistical results that are desired. A t-test, however, is appropriate for the last statistical test concerning the heights and weights of collards and radishes, since both height and weight are quantitative, but also since plant species is binary since there is only two specific levels in this test (collards and radishes). Therefore, a two-sample t-test is adequate to glean statistical conclusions.

RESULTS

There is a statistically significant difference in the total height of seedlings between collards and radishes after the fourth week (P = 0.0013), but not between collards and the radish-collard seed mixture (P = 0.287) or radishes and the radish-collard seed mixture (P = 0.123). This is portrayed in the boxplots that contain the 5-number summary of each species by week. During weeks 1 and 2, the mean height of each species seems similar (Figures 1 and 2).  

Figure 1: A boxplot showing the total heights of each species after week 1 of the

experiment.   

Figure 2: A boxplot showing the total heights of each species after week 2 of the

experiment.

After week 3 of the experiment, the mean heights of collards and the mixture of seeds have a graphically visible difference from the mean height of radish seeds (Figure 3).

Figure 3: A boxplot showing the total heights of each species after week 3 of the

experiment.

This trend continues into week 4, when there is still a visible difference in the mean heights of the 3 species (Figure 4), which is when the P-value of the interaction between the mean heights of the radishes and the collards indicates a statistical difference.

Figure 4: A boxplot showing the total heights of each species after week 4 of the

experiment.

Finally, this trend continues until the end of the experiment (Figure 5), when there is a significant difference between radishes and the radish-collard seed mixture after the fifth week. (P = 0.0214), but there still is not a statistically significant difference in the height of the collards and the height of radish-collard seed mixture (P = 0.6247).

Figure 5: A boxplot showing the total heights of each species after week 5 of the

experiment.

The trend of the lack of difference between the mean heights of the collard seeds and the radish-collard seed mixture is also evident in the boxplot of the heights for week 5 since they are close to each other in size (Figure 5). Additionally, there is no statistically significant difference between the number of seedlings after week 1 (P = 0.89), which is apparent when comparing the 3 boxplots of the seedling counts for the radish seeds, the collard seeds, and radish-collard seeds mixture (Figures 6-8).

Figure 6: A box plot showing the seedling counts of the collard plants, by week.

Figure 7: A boxplot showing the seedling counts of the mixed plants, by week.

Figure 8: A boxplot showing the seedling counts of the radish plants, by week.

There is a statistically significant difference in the number of seedlings after week 5 concerning collards and radishes (P = 0.052) and radishes and the radish-collard seed mixture (P = 0.027), but not between collards and the radish-collard seed mixture (P = 0.678). Between radishes and collards, there appears to be a difference in the number of seeds because the seedling count is 24 in the collard treatments (Figure 6), whereas it is closer to 21 or 22 when analyzing the radish boxplot for week 5 (Figure 8). Comparing the radish boxplot and the mixture boxplot also yields a visual difference since the mixture of seeds appears to be close to 24 seedlings (Figure 7), whereas the boxplot concerning the seedlings after week 5 for the radishes exhibit a number that is much closer to 20 seedlings (Figure 8). Comparing the boxplot of week 5 of the mixture of seeds and the collard seeds, there does not appear to be a visual difference between the 2 of them. The boxplot pertaining to the collards (Figure 6) does not show any striking difference to the mixture boxplot (Figure 7). There also is not a statistically significant difference between final weight and plant species (P = 0.620). The conglomeration of boxplots comparing the final plant mass according to treatment (Figure 9) shows no brazen differences in the means of the plant weights, although treatments 8 and 9 do have a broad interquartile range.

Figure 9:  A boxplot of the plant mass by treatment number.

However, the T-test tests the average plant weight, not the range, which is why a “not significant” result is shown in both the mathematical calculations and the boxplot. Finally, there is a statistically significant difference in average height between collards and radishes (P = 0.003), but there is not a statistically significant difference in average weight between collards and radishes (P = 0.231). The boxplot comparing the average height of the radishes versus the collards shows that the mean height for collards is much greater than the mean height for radishes (Figure 10), whereas the mean weight for radishes is much like the mean weight for collards, which would fail to yield a statistically significant result (Figure 11).

Figure 10: A boxplot of the total height, containing data sets for the radish plants and the

collard plants.

Figure 11: A boxplot of the total weight, containing data sets for the radish plants and the

collard plants.

DISCUSSION

To restate the statistical findings, there was a statistically significant difference between the heights of collards and radishes after the fourth week, and a statistically significant difference between the heights of radishes and the heights of the radish-collard seed mixture after the fifth week.  Also, after the fifth week of the experiment, there is a statistically significant difference between the number of seedlings in collards and radishes as well as the number of seedlings between radishes and the radish-collard mixture.  There is not a statistically significant difference between final plant weight and plant species, nor is there a statistical difference between final plant weight in collards and radishes when considering light exposure too.  Finally, there is a statistically significant difference between the final plant height of collards and radishes when taking the amount of light into account as well. 

At the beginning of this experiment, I hypothesized that collards would grow taller than radishes or the mixture, and that the radishes would grow the heaviest out of the 3 species tested.  Considering my results, the heights of the radishes and collards, as well as the radishes and the radish-collard mixture were proved to be different.  The ANOVA tests are difficult to form conclusions from, because the test can be interpreted as proving that the heights were different; however, it doesn’t prove that the radishes grew taller than the collards or the half-and-half seed mixture.  The raw data makes it easier to make this conclusion, since the creation of the boxplots means the mean of each species is visually represented, and it’s possible to assume that the radishes were shorter than the collards and the radish-collard seed mixture.  Additionally, it was found that there was a statistical difference between the weight of radishes and collards at the close of the experiment.  The T-test proved there is a difference in the weights of radishes and collards, but it did not prove which was heavier, which is again where the boxplot comes in handy.  Comparing the boxplot of the weight of the radishes versus the weight of the collards shows that the radishes were heavier, since the boxplot had a higher mean when pertaining to the radishes.

One of my primary sources was a scientific paper outlining the possible benefits that can result from the intercropping of collards and radishes, and one of the main points was that when done correctly, this intercropping could result in more rational food production as well as increased sustainability.  Another one of my primary sources was a scientific paper that tested light exposure in celery, and the results concluded that a higher amount of UV light or even a longer time being exposed to light can impact the growth of plants positively.  This experiment tested both, which explains why treatment 9 had the second-highest plant mass, and why the total height of the mixture of radish seeds and collard seeds was over 200 centimeters after the fifth week of the experiment.  My final source reflected on the factors that impact root penetration into soil and gave examples as to what soil types are not beneficial for healthy plants, soil that lacked hydration and nutrients.  The ideal soil is a blend between silt and clay: soil that is sparse enough for the roots to penetrate, but still is able to hold water and nutrients for the planted seeds to absorb.  This is why the plants within this experiment had no less than a 79% germination rate (excluding outliers), because the soil was a healthy blend of topsoil and vermiculite, which both aerates soil and retains moisture and nutrients. 

The findings of this experiment can be used to persuade others to use intercropping when they are planting different species of plants, however, radishes and collard greens have somewhat similar needs.  They belong to the same genus, so attempting this experiment again with radishes and a wildly different plant, such as a sunflower (Helianthus genus) would yield different results.  However, this is a valid change that could be made to test if intercropping is good for all botanical species as opposed to those in this specific genus.  It is also interesting to investigate how different this could be with flowering plants like the sunflower, if intercropping would impact pollination at all.  Another improvement would be to lengthen this experiment.  We are under somewhat of a time crunch since the semester is only 16 weeks but starting this project earlier and making it 7-8 weeks could make the differences between the species greater and yield more discoveries.  Ecologically, this experiment gave solid evidence that intercropping in areas can be beneficial, although a flaw in this experiment is that it happened in an enclosed space.  To get a completely random experiment, the outdoors would be preferred, although completing the experiment indoors is a foolproof way to control many variables (temperature, light exposure, climate). 

In conclusion, intercropping is more beneficial than growing crops in one specific place on their own.  The mixture of radish seeds and collard seeds often performed between than its solitary counterparts, and given more time, may have produced healthier offspring.  The melting pot of distinct species could also fight off disease, increasing the viability of the offspring of these mixed plants.  Mixing the seeds of plants that would otherwise be different is an overlooked way to provide these plants with defenses they would otherwise lack and can be repeated in many experiments. 

REFERENCES

Collard greens. Sproutpeople. (n.d.). https://sproutpeople.org/collard-greens/ 

Gwmag. (2023, June 27). Vermiculite: Main uses. What Is Vermiculite? How To Use Vermiculite For Healthier Plants | BBC Gardeners World Magazine. https://www.gardenersworld.com/how-to/grow-plants/how-to-use-vermiculite/ 

Pinheiro, M. de S., Pereira, J. S., Coutinho, C. R., Filgueiras, R. M. C., Guimarães, M. de A., & Mesquita, R. O. (2019, October 3). Intercropping of collard green and radish “cometo”: Spatial arrangement and growing efficiency. Revista Ceres. https://www.scielo.br/j/rceres/a/SQdydGNq9qnkGY995xM7fxt/?lang=en#

Qin, Y., Liu, X., Li, C., Chu, Q., Cheng, S., Su, L., Shao, D., Guo, X., He, Z., & Zhou, X. (2023, December 14). Effect of light intensity on celery growth and flavonoid synthesis. Frontiers. https://www.frontiersin.org/journals/plant-science/articles/10.3389/fpls.2023.1326218/full 

Tomobe, H., Tsugawa, S., Yoshida, Y., Arita, T., Tsai, A. Y.-L., Kubo, M., Demura, T., & Sawa, S. (2023a, May 9). A mechanical theory of competition between plant root growth and soil pressure reveals a potential mechanism of root penetration. Nature News. https://www.nature.com/articles/s41598-023-34025-x#citeas 

Tomobe, H., Tsugawa, S., Yoshida, Y., Arita, T., Tsai, A. Y.-L., Kubo, M., Demura, T., & Sawa, S. (2023b, May 9). A mechanical theory of competition between plant root growth and soil pressure reveals a potential mechanism of root penetration. Nature News. https://www.nature.com/articles/s41598-023-34025-x#citeas 

USDA plants database. (n.d.). https://plants.usda.gov/home/classification/63801