Fostering Students’ Awareness and Attitudes toward River Environments: Development of a Diatom-Based Educational Program

Abstract

This study examined changes in students’ awareness of river environments and the formation of their behavioral attitudes through a diatom-based educational program implemented in secondary schools in India, Japan, and the United States. The program used past and present diatom specimens, river photographs, and the SimRiver simulator to help students understand the relationship between human activities and river water quality. Text mining analysis, using TWINSPAN and Correspondence Analysis, revealed a significant increase in environmental awareness among students after the lessons. The qualitative analysis of students’ responses revealed distinct patterns in their behavioral attitudes. In the Indian groups, a relatively high proportion of students exhibited autonomous and action-oriented attitudes. In contrast, Japanese students frequently showed concern for environmental issues, but their responses often reflected other-dependent thinking. Students in the U.S. group offered fewer behavior-related responses, which may be attributed to the favorable local environment, the geographical disconnect between the lesson materials and the students’ everyday surroundings, and the generally shorter nature of their written responses. The program demonstrated its effectiveness in raising students’ environmental awareness and supporting the development of attitudes toward action. Furthermore, incorporating historical and international comparisons appeared to promote the transition from awareness to action among students. These findings suggest that such lesson programs may serve as valuable contributions to education for Sustainable Development Goal 6 (SDG 6).

INTRODUCTION

Water pollution and sanitation issues are crucial challenges that should be addressed through international collaboration, as advocated by Sustainable Development Goal 6 (SDG 6) (UN-Water, 2023). Raising people's awareness and shaping attitudes toward the water environment is a fundamental factor in solving these problems. The SDG 6 Global Acceleration Framework (UN-Water, 2020) emphasizes the integration of water and sanitation education into school curricula and public outreach programs to facilitate behavioral change related to water conservation and pollution reduction.

There are various hands-on environmental education programs that emphasize the importance of water, implemented both locally (Bae et al., 2011; Fu and Komatsu, 2024; Miyao et al., 2007; Wongchantra et al., 2022; Jeevitnadi, see Websites) and globally (Project WET, see Websites; Project WILD, see Websites). These programs are often conducted as workshops that include fieldwork, games, discussions, etc., primarily as extracurricular activities. Participants in such hands-on programs can observe water in situ, but it is difficult for them to directly understand the causal relationships behind changes in the water environment.

Many Japanese junior high school science and senior high school biology textbooks introduce methods for assessing river water quality using aquatic invertebrates as bioindicators, and/or chemical test kits. This is rooted in the context of Japan's water quality improvement history, where rivers became severely polluted during the period of rapid economic growth in the 1960s but gradually improved following the enactment of the Water Pollution Control Law in 1971. During this process, the former Environment Agency and the Ministry of Construction (now the Ministry of the Environment and the Ministry of Land, Infrastructure, Transport and Tourism) took the initiative in promoting citizen-led water quality assessments using aquatic bioindicators in rivers across the country (Ministry of the Environment; see Websites). The students who participated in the program could learn about the water quality of the surveyed sites based on the environmental adaptations of the observed organisms. However, it was difficult for them to understand the causal relationships behind how the water environment came into being. Thus, they found it challenging to model their observations and grasp the conceptual framework of the relationship between human activities and their resulting impacts on the aquatic ecosystem.

Meanwhile, to help students understand changes in the water environment, computer simulation programs such as Leaf Pack Network Simulation (see Websites), River City (Ketelhut et al., 2010a), EcoMUVE (Ketelhut et al., 2010b), and SimRiver (Mayama et al., 2011; Hoffer et al., 2011; Lee et al., 2011; Lobo et al., 2014) have been developed. With these programs, students can learn about water quality changes and the resulting shifts in aquatic organisms. Generally, since changes in the water environment occur over long periods, it is difficult for students to experimentally observe these processes within the limited duration of school education. However, learning through simulators enables students to model causal relationships by formulating hypotheses and testing them through repeated simulations in a short time. Additionally, collecting and processing data statistically, including numerical analysis and graphing, helps students develop their ability to logically explain the relationship between environmental changes and aquatic organisms, as well as predict future environmental changes. However, one challenge of simulation-based learning is that it often lacks a tangible connection to the real world, making it difficult for students to fully comprehend its relevance to real-world environmental issues (Mayama et al., 2008; Hamed and Aljanazrah, 2020).

SimRiver, which we developed, is an online simulator designed to help students visually understand the relationship between human activities and river water quality through diatoms (Katoh et al., 2004). Diatoms are a fundamental producer in the aquatic ecosystem, responsible for generating approximately 20 to 25% of global photosynthesis, and their name appears in all junior high school science textbooks in Japan. The interface of SimRiver is graphically designed to allow for easy operation. By adjusting environmental factors, users can generate different water conditions, and the corresponding virtual diatom community images are produced based on the given water quality. This simulation program is based on 20 years of accumulated data on water quality and diatom communities that we have collected and analyzed from real rivers (Kobayasi and Mayama, 1989; Mayama, 1999; Satomi et al., 2018). As a result, learners can observe virtual diatom communities as high-quality images, similar to actual diatom samples collected from rivers. With the incorporation of gaming elements, students can actively engage in repeated simulations, naturally acquiring an understanding of the concept of anthropogenic environmental disturbance.

Unlike aquatic invertebrates, which are commonly used as other bioindicators for water quality assessment, diatom specimens do not deteriorate even over centuries because their cells are encased in a glass-like silica shell. Kosakai and Mayama (2015) developed a learning program that combined observations of diatom specimens collected from the same river in the past and present, photographs and a video of the river from both time periods, and a SimRiver simulation. This approach resulted in more realistic responses from students compared to using SimRiver alone, as well as opinions that led to actions for environmental improvement.

To clarify the educational effects contributing to SDG 6, this study implemented lessons in schools in India, Japan, and the United States, utilizing past and present diatom samples, river photographs, and SimRiver simulations. The study aimed to analyze changes in students' awareness of river environments and the formation of their attitudes, and to identify key characteristics of this transformation.

METHODOLOGY

Diatom samples and preparations

Past diatom samples used as learning materials were chosen from the collections of the National Museum of Nature and Science, Japan, the Academy of Natural Sciences of Drexel University, U.S.A., and the Agharkar Research Institute, India. These samples were carefully selected to allow for clear comparison with the present diatom assemblages from the same rivers. Additionally, the present samples were collected from the exact same locations as the past samples (Table 1).

Table 1.Diatom samples used in this educational project

Sample no.	Date	Site	Collector	Source
RM-001847	May 1982	Tama River, Tokyo, Japan	Shigeki Mayama	The National Museum of Nature and Sciences, Japan
GC3525A	Aug. 1948	Lititz Run, Pennsylvania, U.S.A	Ruth Patrick	The Academy of Natural Science of Drexel Univ. U.S.A.
Sr. 4	Mar. 1945	Borivali Stream, Mumbai, India	Hemendrakumar Prithivraj Gandhi	Agharkar Research Institute, Pune, India
M-2133	Aug. 2017	Tama River, Tokyo, Japan	Kengo Satomi	Original, housed at Tokyo Diatomology Lab
M-2339	Aug. 2018	Litiz Run, Pennsylvania, U.S.A	Matthew L. Julius	Original, housed at Tokyo Diatomology Lab
M-2340	Apr. 2017	Borivali Stream, Mumbai, India	Karthick Balasubramanian	Original, housed at Tokyo Diatomology Lab

The samples were heated in a water bath with sulfuric acid containing potassium dichromate for one hour, then rinsed with pure water several times (Mayama, 1993). The suspension containing diatom frustules was dropped onto a cover slip, dried, and then embedded in Mount Media (Fuji Film Wako Pure Chemical, Osaka), a commercial product equivalent to Pleurax (Brown, 1997). Next, the samples were observed under an Axioskop microscope (Zeiss, Oberkochen) equipped with a 63× oil immersion objective lens (N.A. = 1.25).

Images were captured using a DP71 digital camera (Olympus, Tokyo). Subsequently, multiple images were stitched together to create a wide-field composite image. This process was performed for samples from all three countries (Figure 1). For each sample, the original microscope slide along with 99 duplicate specimens has been deposited in the National Museum of Nature and Science, Japan, where they are accessible to third parties upon request.

Figure 1. Photographs of diatom specimens

A and B were collected from the Tama River in Tokyo, Japan, in 1982 and 2017, respectively. C and D were collected from the Borivali Stream in Mumbai, India, in 1945 and 2017, respectively. E and F were collected from Lititz Run in Pennsylvania, U.S.A., in 1945 and 2018, respectively. The diatom communities in A, D, and E are composed exclusively of pollution-tolerant diatoms, whereas B, C, and F exhibit a diverse species composition, including pollution-sensitive species.

Characteristics of the Specimens and Social Conditions of the River Basin at the Time of Collection

The site where the Japanese diatom specimens were collected was a site where a large amount of untreated domestic wastewater flowed in the past. In the diatom community collected at that time, Nitzschia palea, a species classified as a pollution-tolerant diatom (Mayama, 1999), was dominant. The Saprobic Index calculated from the community was 3.7, indicating severe pollution in the water body (Satomi et al., 2018). In contrast, a recent diatom sample collected from the same site indicated good water quality (Saprobic Index = 1.07), demonstrating significant environmental improvement.

The site where the American samples were collected was a small town, but in the past, the river was severely polluted due to wastewater discharge from a chocolate factory located by the river, as well as other industrial facilities (Patrick, 1949; Blankenbiller, 2009). The species composition of the diatom community at that time was similar to that of past samples from Japan. However, a recent sample contained a diverse range of species, reflecting improved environmental conditions.

The river basins where past samples were collected in Japan and the U.S. are now equipped with sewer systems and sewage treatment plants. However, at the time when the past specimens were collected, the sewage treatment system in these areas was still underdeveloped. The species compositions of the past diatom communities clearly reflected the social conditions of those times.

In contrast, the past Indian diatom community consisted mainly of species indicative of good water quality, similar to those found in present-day Japanese rivers. However, the present Indian specimen was dominated by Nitzschia palea, exhibiting a species composition similar to that of past samples from Japan and the U.S.

Present-day Indian society is experiencing severe water pollution due to rapid population growth and an underdeveloped sewage system in urban areas. However, diatom analysis suggests that, before national independence, the river basins were well-preserved in a rich natural environment, as neither rapid urbanization nor industrialization had yet developed.

Digital Learning Resources

There are fewer photographs and videos of polluted rivers compared to clean ones. Even fewer exist for the past. In this study, images of polluted rivers were obtained from the DiatomProject webpage that we developed, which includes historical and contemporary photos and videos of polluted rivers from around the world (see Websites), as well as a video produced by NHK for School (The Polluted Tama River, see Websites).

For clean rivers, we also used recent photographs taken in Japan and the U.S. by the authors, while for India, we used a 19th-century photograph of the Yamuna River from the Old Indian Photos archive (see Websites).

The ecosystem simulator SimRiver (see Websites), which models the relationship between human activities, river water quality, and diatom communities, has been updated annually based on user feedback. In this study, we used version 6 of the program.

Schools and Students

Lessons were conducted between October 2017 and September 2018 at a national high school in Tokyo, Japan, a private secondary school in Bangalore, India, where English was used as the medium of instruction, a public secondary school in Bangalore, India, where Kannada was used, a private secondary school in Pune, India, where Marathi was used, and two public high schools near St. Cloud, Minnesota, USA.

The number of students who participated in the lessons was as follows: 108 in Japan (Grade 10), 112 in the English-medium school in India (Grades 8 and 9), 127 in the Kannada-medium school in India (Grades 8 to 10), 122 in the Marathi-medium school in India (Grades 8 to 10), 34 in one U.S. high school (Grade 10), and 60 in the other U.S. high school (Grade 10). Although the number of students in the two U.S. high schools was relatively small, their backgrounds and feedback were similar. Therefore, the data from both schools were combined for analysis as a single group of 94 students.

From this point forward, the schools will be referred to as JP for the school in Japan, IE for the English-medium school in India, IK for the Kannada-medium school in India, IM for the Marathi-medium school in India, and US for the schools in the U.S.

Lesson implementations and analyses

The lesson program, the flow of which is outlined in Table 2, lasted for two class periods or an equivalent duration at each school. The order in which diatom specimens were introduced differed in each country; specimens collected from that country, both past and present, were shown first, followed by specimens from other countries for comparison. Some of the authors instructed the students in the classes. In Japan, the lessons were conducted in Japanese. In the U.S. and at the IE school in India, they were conducted in English. In the IK and IM schools, the authors’ English instructions were translated into Kannada and Marathi, respectively, by local collaborators.

Table 2.Lesson flow for two class periods

Duration (min)	Activity
First period
10	Pre-survey: Students provide descriptive answers.
8	Questioning: Asking students how they can determine past river water quality. Introduction: Explanation of what diatoms are.
17	Observation: Students observe two diatom specimens without being informed whether they are from the past or present. Discussion: Analyzing the characteristics of the specimens using a diatom shape guide (Appendix 1). Comparison: Introducing photographs of the same river taken in the past and present. Concept Introduction: Explanation of diatoms as bioindicators.
25	Explanation: Demonstration of how to operate SimRiver.
Second period
20	Simulation Activity: Students manipulate environmental settings to generate virtual diatom specimens that resemble the two real specimens, to understand the river basin environment at the time the real specimens were collected. Presentation: Students present their simulation results and share their findings.
15	Discussion: Determining the time period when the two real specimens were collected. Comparison: Introduction of real diatom specimens collected from other countries, along with photographs from the past and present. Questioning: Asking students to determine the time period when each specimen was collected.
10	Post-survey: Students provide descriptive answers.
5	Discussion: Reflection on global river environments.

Before and after the lessons, students were asked the same question: “What do you think about river environments now?” and provided open-ended descriptive answers. The responses were analyzed using text mining methods. Prior to the analysis, answers written in languages other than English were translated into English. Each response was first subjected to morphological analysis, and the words contained in the responses were extracted using the freeware KH Coder 3 (see Websites) by Higuchi (2016; 2017).

Based on the results, a matrix was created in which each row represented a student's response and each column represented a word. This matrix was analyzed using TWINSPAN (Hill 1979), which classified both responses and words into distinct groups. TWINSPAN is a hierarchical clustering method for multivariate data that combines an ordinal structure derived from a procedure similar to the first axis of Correspondence Analysis with subsequent binary classification guided by the selection of indicator species—substituted by indicator words in this study. For further details on TWINSPAN, see Appendix 2. The same matrix was also used for Correspondence Analysis to determine the ordering of the responses.

TWINSPAN and Correspondence Analysis were conducted using PC-ORD (Wild Blueberry Media LLC, Corvallis, USA) and KH Coder 3, respectively. In the qualitative analysis of students’ responses, statistical testing was conducted using the two-proportion z-test with Python’s statsmodels package, and multiple comparisons were adjusted using the Benjamini–Hochberg False Discovery Rate (FDR) correction. For the detailed conditions and program settings used in both analyses, see Appendices 3–5.

RESULTS AND DISCUSSION

Changes in students' awareness

A matrix composed of student’s responses and the words they described was classified using TWINSPAN, resulting in nine response groups (R1 to R9) and eight corresponding groups of words labeled as Human & Pollution, Cause of Pollution and Ecosystem e.t.c. (Figures 2). In the pre-survey, students in all school groups (US, IE, IM, IK) except for the JP group were predominantly classified into R3, which included words from the groups labeled as Current Situation, Human & Pollution, and Causes of Pollution, indicating anthropogenic contamination of rivers, such as garbage. However, the frequency of each word in these groups varied among countries. For example, in Human & Pollution group “clean”, “polluted” and “human” were more frequently used by Indian students, “country” and “pollution” were more frequently described by American students, and "flow," "time," "increase," "small," and "use" were used without bias by students from any particular country.

Figure 2. Relationship between Student Response Groups and Word Groups Classified by TWINSPAN

This figure is divided into two main regions. Region A (left) shows the proportions of each student group that were classified into each response group before and after the lesson. Region B (right) illustrates the associations between response groups and word groups. Within each word group, frequently used words are shown in bold. Words that were characteristically used in only one country are marked with (J) for Japan, (US) for the United States, and (I) for India. Each word group is labeled for clarity. Labels related to river pollution are shown with a dark blue background, while those not directly related are shown with a brown background. Region B also uses black dots to indicate strong associations between response groups and word groups. The dendrograms at the top and left represent the hierarchical clustering of the response groups and word groups, respectively. Number of students (pre-/post-survey): JP (103/109), US (63/73), IE (109/111), IM (121/121), IK (122/126).

Following R3, R4 (more prevalent response in the US and IM) and R5 (more prevalent in the US and IK) were also notable. The students in the R4 mentioned only anthropogenic pollution, whereas those in the R5 also referred to the environment.

In the post-survey, however, the number of responses classified into R3 decreased. Conversely, the number of students classified into R2, which frequently included words from the groups labeled as International/Over-time Comparison and Improvement, Current Situation, and Human & Pollution, increased the most. This suggests that students began to develop ideas for improving river conditions through international and/or temporal comparisons. In the JP school, the proportion of students classified into R2 was already the highest in the pre-survey (30.1%) and increased significantly to 80.7% in the post-survey.

The frequency of word usage in the International/Over-time Comparison and Im-provement group varied significantly by country, indicating differences in students' perspectives on improvement.

In the post-survey, the total proportion of students who described any of R1 to R5 exceeded 97% in all student groups except for the US, demonstrating a significant increase in awareness of water pollution. On the other hand, the proportion of US students describing these responses was 87.7%, which, although relatively high, was lower compared to students from Japan and India. This may be attributed to the relatively well-preserved natural environment around their schools.

The correspondence analysis plot (Figure 3) illustrates the distribution patterns of the vocabulary used in students’ descriptive responses and the relationships between those words and the student groups who provided them. This analysis was conducted using the same response × word matrix as TWINSPAN; however, unlike TWINSPAN, which focuses on classification, correspondence analysis visualizes how each group’s awareness changed relatively before and after the lesson within a shared semantic space.

Figure 3. Correspondence analysis of student responses before and after the lesson

Each red square represents the centroid of a student group’s responses at each time point (pre- and post-survey), and arrows indicate shifts in awareness. The plot is based on the same response × word matrix used in the TWINSPAN analysis. Percentages in parentheses represent the variance explained by each axis.

Abbreviations: Developing C. = developing countries; Developed C. = developed countries; For. C. = foreign countries; W.Q. = water quality; H.A. = human activity; S.T.P. = sewage treatment plant; H.P. = human population; Waste W. = wastewater; W.B. = water bodies; Riv. Env. = river environment; Riv. W. = river water.

Dimension 1 represents the temporal change in students’ awareness, from pre-survey responses (right) to post-survey responses (left), reflecting the conceptual shift induced by the lesson. Dimension 2 reflects students’ orientation toward development and change. The upper area contains words related to improvement, technology, and abstract social concepts, while the lower area includes country names and institutional or present/past-oriented terms. Red squares represent the average positions (centroids) of each group, while the arrows indicate the direction and magnitude of conceptual shifts resulting from the lesson.

Only words that appeared at least 25 times were included in the analysis, and for visualization purposes, the 60 words with the highest chi-square values were selected (Appendix 5). As such, the results reflect the most distinctive awareness patterns associated with each country. Words that appeared frequently both before and after the lesson—particularly those related to river pollution—were concentrated near the origin of the two axes. Words located near each group’s centroid are considered to reflect awareness tendencies characteristic of that group.

Although students expressed a wide range of ideas in their free responses, the TWINSPAN analysis showed that river pollution was a common awareness across all school groups, both before and after the lesson. However, the associations and contexts of that awareness differed by country.

As summarized in Table 3, the words that characterized student groups before and after the lesson varied considerably, clearly reflecting group-specific transformations in awareness. For example, Indian students initially expressed concrete, everyday-life-related concerns, such as the use of river water or associated health risks. After the lesson, their responses included terms related to comparisons across time and between countries, indicating a broader and more reflective awareness. Japanese students, by contrast, initially expressed their awareness of rivers using abstract concepts and words associated with their personal living environment, such as house. After the lesson, they began using terms related to social improvement and development, such as development and water quality. American students initially showed awareness of the natural environment through words such as life and fish, but after the lesson, they used terms suggesting increased global awareness and concrete solutions, such as foreign countries, awareness, and sewage treatment plant.　Examples of how these keywords were used in actual student responses can be found in Appendix 6.

Table 3.Key characteristics of students’ awareness transformation before and after the lesson, by country

	Pre-survey	Post-survey
India	disease, drinking, dangerous, fish	clean, human, India, Japan, past, present, pollution
Japan	organisms, human, important, house	activity, change, development, technology, water quality
U.S.A	fish, life, organisms	awareness, foreign countries, pollution, sewage treatment plant

The correspondence analysis also revealed that the three Indian groups—IE, IK, and IM—showed remarkably similar directional shifts in awareness despite differences in location, school type (public/private), and instructional language. These results suggest that the changes in students’ awareness brought about by the lesson may be fairly generalizable within India. In contrast, whether the distinctive changes observed among students in Japan and the United States are generalizable remains uncertain and requires verification through more extensive comparative studies.

Formation of attitudes

While TWINSPAN and Correspondence Analysis primarily identified students’ awareness toward river environments—especially in the post-survey—as underlying their attitudes, we further analyzed students’ behavioral aspects of attitude through a conventional method: interpreting and classifying their descriptive responses.

In the post-survey, students’ responses included not only descriptions of learning content but also opinions expressing their desire to improve river environments. Such behavior-related descriptions were observed in approximately 60–70% of students in the JP, IE, IM, and IK groups, respectively (Table 4).

Table 4.Types and frequency of students' attitudes in the post-survey

		Student groups (%)
		JP	US	IE	IM	IK
Answers with opinions on improving river environments*		63	20.3	66.7	66.1	72.2
	Non-autonomous and other-dependent thinking*	59.3	8.6	49.5	52.1	50
	Autonomous and action-oriented thinking with specificity**	3.7	11.4	17.1	14	22.2
Answers without opinions on improving river environments*		37	79.7	33.3	33.9	27.8
Number of students		108	79	111	121	126

* Significant differences were found between US and all other groups (p < 0.00001).

** Significant differences were found between JP and all other groups, with p-values ranging from < 0.05 to < 0.001.

Values in bold indicate groups that showed no significant differences from each other in multiple comparisons.

Furthermore, students’ opinions regarding behavioral attitudes were categorized into two types: those they could not carry out themselves and those they could. The former included actions that governments or companies can take, such as providing international aid, constructing sewage treatment plants, enacting laws and regulations, and developing technology. Some of these actions were also partially reflected as group-characterizing words in the TWINSPAN and Correspondence Analysis results (Figures 2 and 3). This category also included non-autonomous, other-dependent thinking, exemplified by statements like “People should do X” or “I expect X to happen,” which may become feasible in the future. This type of thinking was observed in nearly half of the students in the JP, IE, IM, and IK groups. The latter category reflected autonomous, action-oriented thinking, exemplified by not throwing garbage into rivers, advising friends and family against improper waste disposal, and participating in river clean-up activities. While fewer in number compared to the former category, such responses were more frequent in the Indian groups (11–22%) than in the Japanese group (3.7%). Although fewer students in the US group expressed opinions related to behavioral attitudes, those who did often demonstrated autonomous thinking, and the proportion (11%) did not differ significantly from that of the Indian groups (Table 4).

The responses of Indian students reflect an internal motivation shaped by their learning and daily exposure to polluted rivers. In contrast, the limited number of autonomous responses from JP students may be related to the absence of polluted rivers in their residential environment.

As for the US group, the lack of opinions on improving river environments may be attributed not only to the relatively good condition of their local environment, but also to the fact that the visual and biological records of polluted rivers shown during the lesson were from regions different from where they live. Additionally, their responses were generally shorter compared to those of other groups, which may have limited the expression of such opinions. Nevertheless, the fact that more US students expressed autonomous thinking than those in the JP group—and at a frequency not significantly different from that of the Indian groups—may suggest the influence of cultural factors, such as the tendency among Japanese people to be less assertive.

Educational implications and global context

In the context of educational contributions to SDG 6, raising students’ awareness of river environments constitutes the first step, while fostering attitudes toward action represents the second. The survey questions were deliberately kept simple to avoid leading respondents toward particular answers through the wording, thereby ensuring that the responses reflect students’ own thoughts rather than external influences.

The high frequency of responses expressing students’ attitudes toward improving river environments in the JP and Indian groups suggests that the lesson design functioned effectively for these students. In contrast, the results from the US group indicate a need to reconsider the post-lesson activities and instructional materials provided for them.

As noted at the beginning of this paper, a variety of educational activities related to river water have been implemented worldwide. In addition, many countries have provided Official Development Assistance (ODA) to developing nations. However, as indicated in the SDG Progress Report (United Nations Statistics Division, 2024), efforts to address water pollution in developing countries remain insufficient.

Historically, improvements in water quality in many countries have been preceded by heightened public awareness in response to critical environmental conditions. This was followed by governmental and municipal efforts, including the enactment of laws and regulations and the construction of wastewater treatment systems (Bate, 2019; Hartig, 2010; Hill, 2019; JICA, 2022).

When students in developed countries learn about the historical conditions of their own river environments and the processes through which improvements were achieved, and engage in dialogue with students in developing countries about the outcomes and behavioral attitudes involved, such interactions may contribute to the global resolution of water-related issues. Collaborating with local researchers and educators to develop more effective instructional methods and learning environments for this purpose remains an important area for future research. Furthermore, differences in students’ awareness and attitudes may also be influenced by the educational environments in which they are situated. Therefore, identifying the reasons behind these differences will require multifaceted investigation from various perspectives.

CONCLUSION

In this study, we implemented lessons in secondary schools in India, Japan, and the United States using past and present diatom specimens, river photographs, and the SimRiver simulation. These lessons aimed to enhance students’ awareness of river environments and support the development of action-oriented attitudes. TWINSPAN and Correspondence Analysis revealed that students’ awareness shifted significantly after the lessons.

In addition, qualitative analysis of students’ descriptive responses showed that a relatively high proportion of students in the three Indian groups expressed autonomous and action-oriented attitudes toward improving river environments. In contrast, many students in the Japanese group demonstrated other-dependent thinking, although concern for environmental action was still frequently observed.

Students in the United States expressed fewer opinions related to behavior, which may be attributed to their relatively well-preserved local environments, the geographical distance between the lesson materials and their lived context, and the generally shorter nature of their responses. These findings suggest that students’ environmental awareness and the expression of their attitudes are influenced by the social and cultural contexts of their respective countries.

This lesson program was designed as an educational contribution to SDG 6 to raise students’ environmental awareness and support the formation of autonomous and practical attitudes. Its effectiveness was partially confirmed. In particular, the incorporation of comparisons between past and present conditions, as well as international perspectives, may have contributed to the development of students’ awareness into attitudes and actions.

Moving forward, further development of this lesson model and its application to diverse educational settings will require collaboration with educators and researchers across countries and regions. Interactive learning among students from different cultural backgrounds may also help foster a sense of responsibility toward global environmental issues.

ACKNOWLEDGEMENTS

The authors would like to thank M. Uchiyama, M. Olson, J. Dammann, M. A. Khan, S. Reddy, and L. Krishnamurthy for their support in arranging the class activities. We are also grateful to B. Alakananda and N. Wadmare for their assistance with translations, as well as to M. Potapova for providing information on historical American specimens. This work was supported by JSPS KAKENHI Grant Number 23K02759, 19K03113 and 16K01011.

References

•  Bae,  H.-S.,  Choi , D. S. and  Kim , Y. R. (2011) The effects of urban stream ecological experience program on the environmental sensitivity of elementary school students. Korean Journal of Environmental Education 24: 66–75.
•  Bate,  D. (2019) The death of the Delaware River. WHYY. https://whyy.org/articles/the-death-of-the-delaware-river/ <accessed: March 21, 2025>
•  Blankenbiller,  K. (2009) Images of America. Lititz. Arcadia Publishing, Charleston, U.S.A.
• Brown, P. A. (1997) A review of techniques used in the preparation, curation and conservation of microscope slides at the Natural History Museum, London. The Biology Curator 10 Special Supplement: 1–34.
•  Fu  S.-P.  Komatsu  H.K. 2024 Evaluating the impact of place-based education: insights from a river environmental program in Taiwan Journal of International Cooperation in Education 26: 153–170.
•  Gauch , Jr. H. G. and  Whittaker , R. H. (1981) Hierarchical classification of community data. Journal of Ecology 69: 537–557.
•  Hamed,  G. and  Aljanazrah,  A. (2020) The effectiveness of using virtual experiments on students’ learning in the general physics lab. Journal of Information Technology Education: Research 19: 977–996.
•  Hartig , J. H. (2010) Burning river. Revival of four urban-industrial rivers that caught on fire. Multi-Science Publishing Co Ltd., Brentwood, UK.
•  Higuchi,  K. (2016) A two-step approach to quantitative content analysis: KH Coder tutorial using Anne of Green Gables (Part I). Ritsumeikan Social Sciences Review 52(3): 77-91. https://www.ritsumei.ac.jp/file. jsp?id= 325881
•  Higuchi,  K. (2017) A two-step approach to quantitative content analysis: KH Coder Tutorial using Anne of Green Gables (Part II). Ritsumeikan social sciences review 53(1): 137-147. https://www. ritsumei.ac.jp/file. jsp?id=346128
•  Hill,  L. (2019) The Chicago River. A natural and unnatural history. Southern Illinois University Press, Carbondale, U.S.A.
•  Hill , M. O. (1979) TWINSPAN—a FORTRAN program for arranging multivariate data in an ordered two-way table by classification of the individuals and attributes. Ecology and Systematics, Cornell University, Ithaca, New York. https://www.researchgate.net/ publication/240311310_TWINSPAN-A_ Fortran_Program_for_Arranging_Multivariate_Data_in_an_Ordered_Two-way _Table_by_ Classification_of_The_Individuals_and_Attributes#full-text <accessed: March 21, 2025>
•  Hoffer J Mayama S Lingle K Conroy K Julius M.L. (2011) SimRiver, envi-ronmental modeling software for the secondary science classroom. Science Scope 34(5): 29-33.
• JAICA (2022) Japan’s experience in water resources management. Japan International Cooperation Agency. https:// openjicareport.jica.go.jp/pdf/1000047167.pdf <accessed: March 21, 2025>
•  Katoh,  A,  Mayama,  S.,  Ohmori,  H. and  Seino,  S. (2004) Construction of "SimRiver": a simulation program for river epilithic diatom assemblage composition. Japnese Journal of Educational Technology 28: 217–226.
•  Kobayasi,  H. and  Mayama,  S. (1989) Evaluation of river water quality by diatoms. Korean Journal of Phycology 4: 121–133.
•  Kosakai,  K. and  Mayama,  S. (2015) Envi-ronmental education in high school for understanding river environment changing over long periods of time: A class activity using preserved diatom slides, simulation software and video images of an old river. Bulletin of Tokyo Gakugei University. Division of Natural Sciences 67: 33–34.
•  Ketelhut , D.  J. K.,  Clarke, J. and  Nelson , B. C. (2010a) The development of River City, a multi-user virtual environment-based scientific inquiry curriculum: Historical and design evolutions. In: Jacobson, M. J. and Reimann, P. (eds.) Design for Learning Environments of the Future. International Perspectives from the learning sciences, pp. 89-111. Springer Science + Business Media. Berlin, Germany.
•  Ketelhut , D. J. K.,  Nelson , B. C.,  Clarke,  J. and  Dede,  C. (2010b) A multi-user virtual environment for building and assessing higher order inquiry skills in science. British Journal of Education Technology 41: 56–68.
•  Lee , J. H.,  Cheong,  C.,  Kwon , N. J.,  Kim , Y. J.,  Park , H. G.,  Mayama,  S.,  Katoh,  K. and  Omori,  H. (2011) Trial of educational computer simulation software ‘SimRiver’ for assessment of river water quality for environmental education in schools. Environmental Education, Korea 24(1): 40–48.
•  Lobo , E. A.,  Wetzel , C. E.,  Schuch,  M. and  Ector,  L. (2014) Diatomáceas epilíticas como indicadores da qualidade da água em sistemas lóticos subtropicais e temperados brasileiros. EDUNISC, Santa Cruz do Sul, Brazil.
•  Masuda,  Y.,  Katoh,  K.,  Murakami,  A. and  Watanabe,  S. (2001) A study on urban open space for environmental education based on questionnaires to teachers. Journal of The Japanese Institute of Landscape Architecture 64: 627–631.
•  Mayama,  S. (1993) Eunotia sparsistriata sp. nov., a moss diatom from Mikura Island, Japan. Nova Hedwigia Beiheft 106: 143–150.
•  Mayama,  S. (1999) Taxonomic revisions to the differentiating diatom groups for water quality evaluation and some comments for taxa with new designations. Diatom 15: 1–9.
•  Mayama,  S.,  Katoh,  K..,  Ohmori,  H.,  Seino,  S.,  Kokubuda,  K., and  Oshikata,  K. (2008) Trial and evaluation of "SimRiver": Educational simulation software for assessment of river water quality using diatoms. Japanese Journal of Biology Education 48: 10–20.
•  Mayama,  S.,  Katoh,  K.,  Omori,  H.,  Seino,  S.,  Osaki,  H.,  Julius,  M.,  Lee , J. H.,  Cheong,  C.,  Lobo , E.  A..,  Witkowski, A.,  Srivibool,  R.,  Muangphra,  P.,  Jahn,  R.,  Kulikovskiy,  M..,  Hamilton , P. B.,  Gao , Y. H., and  Ector,  L. (2011) Progress toward the construction of an international web-based educational system featuring improved "SimRiver" for understanding of river environments. Asian Journal of Biology Education 5: 2–14.
•  Miyao,  H.,  Kamata,  T.,  Yoshino,  H. and  Tsuji,  M. (2007) Support for environmental learning using rivers as a theme. Comprehensive Report of the River Environment Research Institute 13: 1–11.
•  Patrick,  R. (1949) A proposed biological measure of stream conditions, based on a survey of the Conestoga Basin, Lancaster County, Pennsylvania. Proceedings of the academy of Natural Sciences of Philadelphia 101: 277–341.
•  Satomi,  K.,  Mayama,  S. and  Saito-Kato,  M. (2018) Water quality change indicated by past and present diatom samples collected from Tama River: Basic research toward the development of educational program for understanding relationship between human activity and river environment change using diatoms. Bulletin of Tokyo Gakugei University. Division of Natural Science 70: 31–53.
•  UN-Water  (2023) The Sustainable Development Goal 6 Synthesis Report on Water and Sanitation Global Acceleration Framework. UN-Water. https://www.unwater.org/sites/ default/files/2023-08/UN-Water_SDG6_ SynthesisReport_2023.pdf <accessed: March 21, 2025>
•  UN-Water  (2020) The Sustainable Development Goal 6 Global Acceleration Framework https://www.unwater.org/sites/default/files/app/uploads/2020/07/Global-Acceleration-Framework.pdf <accessed: March 21, 2025>
• United Nation Statistics Division (2024) SDG Report 2024 Goal 6. https://unstats.un.org/ sdgs/report/2024/Goal-06/ <accessed: March 21, 2025>
•  Wongchantra,  P.,  Ongon,  S,  Junkaew,  L. and  Sookngam,  K. (2020) The effect of environmental education learning for enhancing rivers management in northeast Thailand. Journal of Educational Issues 8: 152–173.
• WEBSITES
• DiatomProject https://www2.u-gakugei.ac.jp/~diatom/ index.html <accessed: June 19, 2025>
• Jeevitnadi https://www.jeevitnadi.org/ <accessed: June 19, 2025>
• KH Coder 3 https://khcoder.net/en/ <accessed: June 19, 2025>
• Leaf Pack Network Simulation https:// leafpacknetwork. org/virtual-stream-study/ <accessed: June 19, 2025>
• Ministry of the Environment. Revision of the Simple Water Quality Survey Method by Aquatic Organisms - Unification of Survey Methods of the Environment Agency and the Ministry of Construction-. https://www. env.go.jp/press/943.html?utm_source=chatgpt.com <accessed: June 19, 2025>
•  NHK for  School. The Polluted Tama River https://www2.nhk.or.jp/school/watch/clip/?das_id=D0005311065_00000 <accessed: June 19, 2025>
• Project WET https://www.projectwet.org/<accessed: June 19, 2025>
• Project WILD https://www.fishwildlife. org/projectwild <accessed: June 19, 2025>
• SimRiver http://lbm.ab.a.u-tokyo.ac.jp/%7Eomori/ srhtml/diatom.htm <accessed: June 19, 2025>
• Yamuna River from the Old Indian Photos
• https://www.oldindianphotos.in/2016/01/the-railway-bridge-
• on-river-yamuna.html <accessed: June 19, 2025>

Appendices

Appendix 1. Diatom shape guide

Appendix 2: TWINSPAN

TWINSPAN (Two-Way Indicator Species Analysis) is a binary hierarchical classification method originally developed by Hill (1979) for the purpose of analyzing community structures in ecology. This method is characterized by the combination of an axis extraction procedure similar to Correspondence Analysis to capture the ordinal structure of the data, and binary classification based on the occurrence patterns of indicator species. Gauch and Whittaker (1981) systematically evaluated the validity of TWINSPAN through comparisons with other hierarchical classification techniques.

In TWINSPAN, a species × sample matrix is first subjected to an ordination procedure (not a full Correspondence Analysis, but a comparable method internally implemented in TWINSPAN) to extract an underlying gradient structure among the samples. Based on this structure, the samples are divided into two groups, and representative indicator species are then selected for each group. Indicator species are defined as those that frequently appear in one group but rarely in the other, thereby characterizing the ecological features of each group. This division and selection of indicator species are applied recursively, resulting in a binary tree hierarchical structure. The algorithm offers both classification and ordination capabilities and is practical due to its relatively low computational demands, making it suitable for large datasets.

Application in Educational Survey Analysis

This method is not limited to ecology but has also been applied to open-ended questionnaire analysis in the field of education. For example, Masuda et al. (2001) applied TWINSPAN to teachers' written responses regarding urban green spaces, classifying the responses based on lexical occurrence patterns and extracting indicator words. In educational applications, each written response is treated as a sample, and nouns, adjectives, and other selected parts of speech extracted through morphological analysis are regarded as species. A frequency matrix of words × responses is then constructed, to which TWINSPAN is applied. This allows for the hierarchical classification of written responses according to tendencies in word usage.

Furthermore, TWINSPAN's classification originates from a potential structure extracted through correspondence analysis-like processing based on vocabulary appearance frequency. That is, by numerically arranging responses (samples) according to their vocabulary distribution patterns and then classifying them into two groups along this arrangement, differences in the substantive direction of the descriptions can be visualized. During classification, words that characteristically appear in each group are extracted as “indicator words.” These words, analogous to indicator species in ecology, can be interpreted as keywords that symbolize distinctive awareness or cognitive frameworks within each group. Thus, TWINSPAN can serve as an effective tool for statistically and semantically capturing the structural diversity embedded in open-ended survey responses.

Appendix 3. List of synonyms used for analysis

Words described by students	Converted word for analysis
dirty, bad, worse, not clean, unclean	polluted
good, not polluted, better, neat, fresh	clean
other countries, another country	foreign countries
sewage channel, sewage system, drainage system	sewage treatment plant
older days, before, previous days, 1960’s	past
dust, trash	garbage
vanish	disappear
nowadays, now, today, modern, 2010’s	Present
spoil, contaminate,	pollute
ill, sick	disease
harmful, hazardous, unsafe	dangerous
conserve, maintain	preserve
household	domestic
living things, creature	organisms
condition	situation
my country, our country	Japan (for Japanese students), India (for Indian students), U.S.A. (for American students)

Appendix 4. Selection of Words used for TWINSPAN and Correspondence Analysis

Parts of Speech
Noun, pronoun, freign words, adjective.
Extracted compounds	Excluded words
alien species, class activity, COD, developed countries, developing countries, foreign countries, high economic growth, human activity, human population, native species, river environment, river water, Saprobic Index, sewage treatment plant, SimRiver, U.S.A., waste water, water bodies, water pollution, water quality.	able, anything, area, do, due, etc, few, Get, glad, have, kind, less, lot, make, many, more, most, much, other, river, such, thing, think, thought, want, Water, way, work, year.

Appendix 5. Program setting

PC-ORD	KH-Coder
Filter words by Term Frequency: Min. TF = 20 Cut levels: 0.0000 Minimum group size for division = 5 Maximum number of indicators per division = 5 Maximum number of species in final table = 200 Maximum level of divisions = 6	Filter words by Term Frequency: Min. TF = 25 Filter words by Document Frequency: Min. DF = 1 Document Unit: Sentence Filter words by chi-square value: Top 60

Appendix 6. Selected Qualitative Responses from Pre- and Post-Surveys on River Environments

This appendix presents selected qualitative responses from students in India, Japan, and the U.S.A. to the question “What do you think about the river environment?” These responses were collected before and after the lesson and show considerable variation in both length and style of expression. The examples were chosen because they reflect characteristic words commonly used in each student group (Table 3) and were incorporated into subsequent analyses such as TWINSPAN and Correspondence analysis.

	Pre-survey	Post-survey
India	When we observe the current situation, we see that diseases are spreading in India. We need to change this situation. A plan must be developed to minimize the spread of such diseases. Rivers are a primary source of drinking water. To reduce disease, we must keep our rivers clean.	Water is essential for all organisms. Water pollution is increasing due to the growth of the human population, and rivers in India are more polluted than those in other countries. In developed countries like Japan, rivers were more polluted in the past, but now they have become cleaner thanks to the construction of sewage treatment plants. In contrast, rivers in India were clean in the past, but are now polluted due to many changes. As a result, organisms in the river, such as fish, are decreasing. Therefore, to control this pollution, we should build sewage treatment plants.
	Rivers are polluted now, and fish are dying. If we play or swim in the rivers, the pollution can cause diseases in us.	We should protect rivers by avoiding the dumping of waste, plastics, and waste oils into them. Humans should change their attitude toward cleanliness. In the past, rivers in India were kept clean, and no waste was thrown into them. However, with increasing pollution, rivers have become contaminated and are no longer suitable for drinking. Similarly, in Japan, rivers were polluted in the past, but now people understand the importance of keeping rivers clean. If we continue to dump waste into rivers, the animals, plants, and other organisms living in the water will be harmed. That is why everyone says we should protect our rivers.
	The present situation is that all rivers are polluted. The water in the rivers is also contaminated. It contains many dangerous chemicals and gases that are harmful to us. This water is unsafe for drinking. Due to the addition of garbage, these water bodies have become polluted, and as a result, the number of fish has decreased. If this situation continues for a few more years, all living organisms may eventually disappear.	India had clean rivers in the past, but now they are becoming polluted due to increased water pollution and the growing human population near rivers. In the past, Japan, the U.S.A., and Germany faced similar problems. Now, developed countries have recognized the importance of clean rivers, and by enforcing strict laws and maintaining cleanliness, they have managed to improve river conditions. Similarly, we too can clean our rivers and keep them beautiful.
	At present, rivers are more polluted because we are destroying them. I like to play in the river. In the past, river water was clean and safe to drink, but now we cannot drink it because it is polluted. Also, we can no longer catch fish, crabs, and other aquatic creatures.	It is possible to clean river water by finding new solutions and conducting experiments. Developed countries like Japan and the U.S.A. have made efforts to clean their rivers and have been successful. They have stopped dumping garbage and releasing waste into rivers. They have also raised awareness about the importance of rivers, especially for people living in cities near them. In the past, all rivers in India were clean, but now they are polluted. Similarly, rivers in Japan were once heavily polluted, but the current situation is improving.
Japan	When I was a child, I often caught crayfish in a small river near my grandparents' house and watched the flowing water. That river had the clean water I long for, but such rivers are rare in Tokyo or Saitama. More people should care about water quality.	The river conditions in Japan during its period of rapid economic growth and in developing countries today are similar. However, past experiences have improved the river conditions in Japan, and we must develop technologies to keep the river environment clean. As a way to understand the river environment, organisms such as diatoms are useful and very interesting. Through this class activity, students will learn that human activities and the river environment are closely related in a practical way.
	Due to changes brought by human activity and development, river ecosystems have changed, but I do not have a real sense of it myself.	The river environment has changed dramatically from the past to the present. Most of these changes are due to human activity. Humans have created a large amount of polluted water, and the river environment has become contaminated almost without our realizing it. Humans need to understand the river environment better than any other species. We also have the technology to improve it. Japan has succeeded in making its rivers cleaner, and this effort should be extended to other countries as well.
	Rivers are unique and host many different and diverse organisms. Studying all the rivers would be almost overwhelming.	Through the class activities, I learned that human activity is a major factor affecting water quality. Humans use energy to obtain water and build advanced sewage treatment plants and water purification facilities. While humans affect water, water also has a greater impact on the Earth. The water that connects humans to the Earth must continue to be conserved.
	Rivers usually don’t have a strong connection to my daily life, but the Kanda River is near my house. Upstream near the highway, the water is polluted ? we can’t see the bottom and it smells bad. I wonder what its original condition was. Clean water supports many organisms, but people pollute it, and that saddens me.	Rivers are originally natural resources, and humans have respected nature. People used to clean rivers, go fishing, and appreciate nature’s life. However, due to development, humans have become foolish. They pollute rivers with domestic wastewater, kill fish, and harm themselves. Currently, developed countries have a sense of crisis and are trying to restore river environments to their original state through technology and effort. We can keep rivers clean through our daily activities. However, developing countries are becoming polluted like the rivers in developed countries’ past. It is important for developed countries to cooperate with others to clean rivers.
U.S.A.	A lot of rivers are becoming overcrowded with both visitors and fishermen, so humans tend to leave garbage, harm fish, and litter the shores. As a result, the fish population is decreasing, especially the smaller ones. Rivers need to be kept clean.	I was surprised to learn that many rivers today, such as those in the U.S.A. and Japan, have cleaner water quality than in the past. I think rivers in developing countries like India still have very polluted water, and countries in similar situations should take action to change this. I believe making such changes would improve the quality of life in many developing countries.
	I think rivers are cool because there are plenty of different kinds of fish and other types of wildlife in and around them.	I think that rivers are still quite polluted, and obviously, some places are cleaner than others. The biggest threat to clean and safe water is humans, so areas with many people are usually more polluted.
	I think rivers are really great habitats for fish and other organisms. I believe we need to work on reducing the pollution we are putting into our rivers.	Rivers today are improving in quality due to greater awareness and better sewage treatment facilities. I think we should continue to raise awareness.
	I like to go to rivers and see God’s glorious creation. There are so many insects and animals that live around rivers. Sometimes, if the river is clean enough, fish can swim through it. Some rivers are polluted, which is very sad because rivers provide so much for us—transportation, wildlife, and more.	I think rivers still need a lot of work, and we can help by advancing technology and funding research on sewage treatment and filtration for rivers and streams. I believe that India and the Philippines need to improve their filtration technology to preserve their streams.