AI-Enhanced Design and Application of High School Geography Field Studies in China: A Case Study of the Yellow (Bohai) Sea Migratory Bird Habitat Curriculum

Abstract

China’s Yellow (Bohai) Sea bird habitat is an important ecological region. Its unique ecology and challenges provide rich resources for research and study. Our course design concept is supported by AI technology, and improves students’ abilities through innovative functions such as dynamic data support, personalized learning paths, immersive research and study experience, and diversified evaluation mechanisms. The course content revolves around the “human–land coordination concept”, including pre-trip thinking, research and study during the trip, and post-trip exhibition learning, covering regional cognition, remote sensing image analysis, field investigation, and protection plan display activities. ERNIE Bot participates in optimizing the learning path throughout the process. The course evaluation system starts from the three dimensions of “land to people”, “people to land”, and the “coordination of the human–land relationship”, adopts processes and final evaluation, and uses ERNIE Bot to achieve real-time monitoring, data analysis, personalized reports, and dynamic feedback, improving the objectivity and efficiency of evaluation, and helping students and teachers optimize learning and teaching. However, AI has limitations in geographical research and study, such as insufficient technical adaptability, the influence of students’ abilities and habits, and the adaptation of teachers’ role changes. To this end, optimization strategies such as improving data quality and technical platforms, strengthening student technical training, enhancing teachers’ AI application capabilities, and enriching AI functions and teaching scenarios are proposed to enhance the application effect of AI in geographical research and promote innovation in educational models and student capacity building.

1. Introduction

In recent years, the rapid development of AI technology has brought unprecedented changes to the field of education [1]. With the support of cutting-edge technologies, such as generative AI, natural language processing (NLP), and educational data mining, the traditional teaching model has gradually been replaced by a digital and intelligent educational ecosystem [2]. In recent years, the Ministry of Education has repeatedly emphasized the deep integration of information technology in the field of education, and has proposed to give full play to the potential of AI in curriculum design, teaching evaluation, and resource management to meet the needs of students’ personalized learning and all-round development [3]. These policies have pointed out the direction for AI-enabled education and have provided a theoretical basis and technical support for the innovation of specific subject teaching [4]. The application of AI technology in education is not a sudden rise, but the result of long-term exploration [5]. From early online learning platforms to today’s intelligent teaching systems, AI has gradually penetrated into all aspects of education [6]. For example, AI-based learning management systems can dynamically analyze students’ learning behaviors and generate personalized learning paths [7]; educational chatbots can instantly answer students’ questions inside and outside the classroom and provide efficient interaction [8]. This intelligent teaching method with generative AI as the core can not only improve the efficiency of educational resource utilization, but also meet the diverse and personalized learning needs of students [9]. However, although AI technology has been widely used in many fields of education, its application in geography education is still insufficient [10]. As a subject that is both scientific and practical, high school geography involves complex spatiotemporal data analysis, ecosystem simulation, and dynamic research on regional human–land relationships [11]. The traditional teaching model emphasizes teaching geographical theory in class, and students learn through textbooks and test their results through examinations [12]. However, this one-way transmission model makes it difficult for students to truly understand the complexity of “human–land relationships”, especially the connotation of “human–land coordination”, a core aspect of geographical literacy [13]. Geographical education urgently needs a more efficient and participatory teaching model, and the introduction of AI technology provides the possibility of achieving this goal [14].

A geography study tour is an innovative education model that combines theory with practice. It brings students from the classroom into real geographical situations and allows them to experience the dynamic changes in geographical phenomena and the relationship between people and land [15]. Since the Ministry of Education issued the “Opinions on Promoting Study Tours for Primary and Secondary School Students” in 2016, study tours have gradually become an important part of basic education [16]. In particular, the promulgation of the “Geography Curriculum Standards for Ordinary High Schools” in 2017 further clarified the important role of geography study tours in cultivating students’ core literacy [17]. Study tours can not only help students understand the complex relationship between the natural and social environment, but also cultivate students’ ability to solve practical geographical problems and environmental protection awareness through hands-on practice [18]. In the implementation of geography study tours, the “human–land coordination concept” as the key objective of the core literacy of the geography subject has a particularly prominent educational value [19]. The “human–land coordination concept” refers to the harmonious development attitude and concept that humans should uphold when dealing with the relationship with the natural environment [20]. This core literacy emphasizes students’ understanding of the interaction between the natural environment and human activities, and cultivates students’ ability to solve practical problems in complex geographical situations [21]. Numerous studies have shown that geography study tours are of great significance for improving students’ geographical practical ability and comprehensive literacy [16]. For example, some studies focus on the organization and implementation of study tours, emphasizing that through field trips students can gain a deep understanding of the natural and human geographical environments and enhance their perceptual understanding of geographical knowledge [11,12]. Some scholars have explored the positive role of study tours in cultivating students’ regional cognition and geographical practical ability and have found that students can better master geographical survey methods and improve their ability to observe and analyze geographical phenomena during field research [20]. However, there are still some shortcomings in the traditional geography study tour course research. On the one hand, the course design often lacks systematicity and scientificity, relies too much on teacher experience, and finds it difficult to meet the diverse needs of students. On the other hand, the evaluation system is not perfect, focusing more on the students’ knowledge mastery, and ignoring the comprehensive evaluation of students’ thinking ability, innovation ability, “human–land coordination”, and other core aspects of literacy [21]. In addition, in terms of teaching methods, traditional methods are mostly used in a manner that is lacking in terms of the effective use of modern information technology, and it is difficult to fully stimulate students’ interest and initiative in learning. These shortcomings restrict the further development of geography study tour courses and also highlight the need to introduce new technologies for innovation [22].

Integrating AI technology into geography study tours provides a new idea for the innovation of this teaching model [23]. The advantages of AI technology lie in its powerful data processing capabilities and personalized learning support [14]. In geography study tours, AI can empower teaching through data visualization and dynamic analysis, personalized learning support, and real-time feedback and evaluation [14]. Through the above methods, AI-enabled geography study tours can help students analyze geographical phenomena and the relationship between people and land from multiple perspectives, and can help improve their comprehensive thinking and innovation capabilities [24]. In addition, the introduction of AI has greatly reduced the burden on teachers in the process of course design and implementation, allowing them to focus more on achieving teaching goals [25].

We integrated AI technology into a high school geography research course and used the generative AI (ERNIE Bot) developed by Baidu in China to explore how to use AI to empower traditional teaching models. The Yellow Sea (Bohai) Sea migratory bird habitat in China is an important node on the East Asia-Australasian migratory bird flyway with extremely high ecological value. However, the region has long faced ecological challenges such as wetland degradation and biodiversity decline. In this context, the goal of geography research travel is not only to help students understand the urgency of wetland protection, but also to enable students to participate in data analysis and ecological decision-making in real time through the support of AI technology. We aim to explore the specific application of AI technology in course design, implementation and evaluation by building an AI-driven geography research travel model. The following content of this article will be developed around the following: Section 2 introduces the research area and design concepts, explains the significance and challenges of the Yellow Sea (Bohai Sea) migratory bird habitat, and the principles of AI-based course design, such as using ERNIE Bot to integrate the “human-land coordination concept” under the principle of problem-oriented, to provide a framework for teaching. Section 3 talks about AI-assisted course design, including course goals such as using AI to improve students’ various abilities, course content from the three dimensions of “human-land coordination concept”, combining a variety of activity tools to optimize learning paths, and help students explore and innovate. Section 4 builds an evaluation system, based on the three dimensions of “human-land relationship”, details process and final evaluation, and the role of AI in it, to achieve objective, efficient, personalized and dynamic evaluation, and to help optimize teaching. Section 5 analyzes the limitations of AI application, such as problems in technology, students, and teachers, and proposes optimization strategies, such as building a data platform to improve the effect. Section 6 summarizes the results, emphasizes the effect of AI-supported course design and the significance of the evaluation system, outlines the innovation points and practical significance, and looks forward to the prospects of AI application.

2. Study Area and Design Concept

2.1. Study Tour Project Area

China’s Yellow Sea (Bohai) Sea migratory bird habitat is an important node in the field of global ecological protection, located in the center of the East Asia-Australasian Migratory Bird Flyway [26]. This area is famous for its unique natural ecological environment and rich biodiversity, providing a key place for many endangered species to live, reproduce and migrate [27]. As an important part of the world’s natural heritage, the ecological protection of the Yellow Sea (Bohai) Sea migratory bird habitat is not only of global significance, but also an important support for regional economic and social development [28].

As shown in Figure 1, China’s Yellow (Bohai) Sea migratory bird habitat covers coastal areas such as Yancheng, Jiangsu, which are famous for their rich tidal wetland resources. It has large-scale intertidal mudflats and is one of the largest wetlands in East Asia. Every year, more than 2 million migratory birds stop, forage, or breed on this migration route, including rare species such as red-crowned cranes and black-headed gulls. These migratory birds not only rely on wetland resources here, but also play a vital role in global biodiversity conservation. Wetlands are a core component of migratory bird habitats and an important part of the Earth’s ecosystem. However, due to human activities such as reclamation, industrial development, and pollution, the area and ecological functions of the Yellow (Bohai) Sea wetlands have declined significantly in the past few decades. In recent years, through international cooperation and policy promotion, the protection of China’s Yellow (Bohai) Sea migratory bird habitats has achieved remarkable results, which provides students with a realistic and vivid case to understand the relationship between humans and land.

The ecological protection of the Yellow (Bohai) Sea migratory bird habitats provides rich teaching resources and practical opportunities for geography research. During the research activities, students can observe the habitat behavior of migratory birds through field visits to the wetland environment and understand the dynamic changes in the wetland ecosystem. In addition, the conflict between wetland protection and development is also an important geographical issue, which can guide students to think about how to balance the relationship between economic development and ecological protection, thereby deepening their understanding of the “coordination of humans and land”.

2.2. Design Concept

The “coordination of humans and land” is the key objective of the core literacy of geography, emphasizing the harmonious coexistence of human activities and the natural environment [29]. In high school geography teaching, the traditional classroom model makes it difficult for students to truly experience the complexity of the interaction between nature and humans, while geography study tours provide a new teaching method [30]. When designing study courses, integrating the “coordination of humans and land” into the teaching objectives and helping students internalize this concept through specific activities is the core challenge of course design [31].

We proposed a research-based course design concept supported by AI technology, as shown in

Table 1. Comparison of innovative functions and applications of AI technology in geographical research.

Category	AI Technology Capabilities	Specific Applications
Dynamic data support	Process and analyze complex geographic data in real time to provide students with dynamic learning resources.	1. Analyze wetland ecological changes and migratory patterns of migratory birds.
		2. Generate ecological change models based on historical data to help students predict future trends.
Personalized learning paths	Provide personalized learning topic recommendations and adjust task difficulty and content based on student feedback.	1. Students can choose learning topics based on their interests (e.g., migratory bird ecology or wetland protection).
		2. Adjust task content in real time to meet the learning needs of different students.
Immersive study experience	Generate virtual study scenarios and support student data analysis and conclusion generation in real time.	1. Generate a 3D wetland model in the pre-trip stage to help students familiarize themselves with the characteristics of the area.
		2. Provide real-time data analysis support during the in-service stage to accelerate students’ understanding and application of collected data.
Diversified evaluation mechanism	Quantitatively evaluate student performance and generate learning reports to improve learning strategies.	1. Quantitatively evaluate students’ knowledge, practical ability, and innovative thinking.
		2. AI-generated reports help students and teachers improve their learning processes and strategies.

. The core of this design concept is to integrate the problem situation setting, learning the resource optimization and dynamic feedback mechanism in the research-based process through the empowerment of AI, so as to maximize the teaching effect. Compared with traditional research-based design, the addition of ERNIE Bot has brought a lot of innovation to course design.

2.3. Design Principles

To achieve the above innovations, this course design follows the following principles:

1. Problem-oriented: Driven by real ecological problems, guide students to conduct in-depth discussions on the conflict between wetland protection and development. Set up a multi-level problem chain through the ERNIE Bot dialog box, gradually guide students from phenomenon observation to data analysis, and finally propose solutions.

2. Task-driven: Design a variety of learning tasks in the course, such as remote sensing image analysis, bird migration data processing, etc., and use ERNIE Bot’s image and document reading functions to support students in completing tasks.

3. Interactive learning: Enhance students’ participation and interactivity in the learning process through ERNIE Bot’s chatbot and virtual assistant.

4. Collaborative learning: Guide students to work in groups, use ERNIE Bot to share data and assign tasks, and complete complex learning tasks in a team.

By integrating AI technology into geographical research design, students can not only deepen their understanding of the “human–land coordination concept”, but also improve their abilities in many aspects in practice. The first is analytical ability. With the help of AI, students can extract key information from large amounts of data and form scientific judgments. The second is innovation ability. The ecological simulation supported by ERNIE Bot allows students to explore multiple possible solutions and cultivate their creative thinking. Finally, it is about action. After the study, the protection plans and practical suggestions proposed with the assistance of ERNIE Bot help to transform the learning results into practical actions.

3. Geography Research Course Design Assisted by AI

3.1. Course Objectives

The introduction of AI technology has brought unprecedented innovation momentum to geography research courses [32]. It not only promotes the transfer of knowledge, but also provides new perspectives and methods for the cultivation of students’ abilities. Through the deep involvement of AI, this course aims to improve students’ geographical practice ability, systematic thinking and innovation ability, and enhance their awareness of sustainable development [33]. First, the course will use ERNIE Bot to help students collect, analyze, and visualize data, thereby improving their ability to use geographical tools. For example, in the process of formulating and optimizing ecological protection plans, ERNIE Bot will assist students in completing specific tasks, helping them to improve their understanding and application of geographical phenomena and data processing in actual operations. Secondly, the course encourages students to develop systematic thinking and innovation capabilities [3]. With the help of ERNIE Bot to analyze the dynamic changes in the relationship between humans and land in wetland ecosystems, students will think about problems from a holistic perspective, understand complex ecological processes, and propose innovative ecological management plans on this basis [34]. In addition, the scenario simulation and decision support tools provided by ERNIE Bot will stimulate students’ innovative potential when facing real problems [25]. Finally, the course is committed to cultivating students’ awareness of sustainable development [26]. Through the dynamic feedback mechanism and situational learning assisted by ERNIE Bot, students will deeply understand the importance of sustainable development while exploring the complex relationship between wetland protection and human development. During the task implementation and evaluation process, students will be guided to think about the harmonious coexistence of humans and nature, thus forming a comprehensive understanding of ecological protection and rational use of resources.

3.2. Course Content

Based on the core geographical literacy of “the view of human–land coordination”, the course content is developed in three dimensions: “land to people”, “people to land” and “human–land coordination”. Combined with various practical activities such as field observation, data collection, and ecological monitoring, students can explore the relationship between people and the environment in real situations and propose solutions to practical problems with innovative thinking. The design of the research and study program follows the three major links of pre-trip thinking, on-trip research and study, and post-trip learning, ensuring that students can systematically and deeply master the course content in terms of the learning of theoretical knowledge, participation in practical activities, and reflection and summary. Each module is deeply integrated with AI technology to optimize students’ learning path and task completion effect. The specific tasks of each link are carried out through a variety of activities and tools, such as maps, remote sensing images, and data analysis tools, so that students can explore the ecological value of China’s Yellow (Bohai) Sea migratory bird habitat from multiple perspectives, and think about and innovate on its existence in combination with human activities. The on-trip research and study link is the core part of the course. Through field investigation activities in China’s Yellow (Bohai) Sea migratory bird habitat, students can personally experience the interaction between humans and nature and explore the complexity of wetland ecosystems in dynamic changes. The specific content design of the specific study tour course is shown in

Table 2. Course design for the study tour to the Yellow Sea (Bohai Sea) bird habitat in China.

Modules	Activity Topics	Place	Activity Content	How to Get Involved with AI	Design Intent
Thinking before going	Regional cognition and background analysis	Classroom	1. Use AI-generated dynamic maps and wetland models to learn about migratory bird routes and wetland change history.	1. ERNIE Bot generates dynamic maps and wetland models, providing intuitive learning resources.	Guide students to understand the dynamic changes in wetland ecology and its global conservation significance, laying the foundation for subsequent field investigations.
			2. Use educational chatbots to answer questions related to ecological protection and sort out the relationship between wetlands and migratory bird ecosystems.	2. ERNIE Bot’s chatbot supports students in answering questions and supplementing knowledge in real time.
	Remote sensing images and environmental monitoring	Classroom	1. Students use AI-assisted remote sensing images to compare changes in wetland area over different eras.	1. ERNIE Bot processes remote sensing images over multiple time periods and generates wetland change trend maps in real time.	Cultivate students’ ability to analyze geographic information and help them understand the driving factors of wetland ecological changes.
			2. Use AI to analyze the specific impact of human activities on wetland ecology.	2. ERNIE Bot automatically identifies key impact points of human activities (such as industrial development and tourism) on the ecosystem.
Study on the Go	From the edge to the center: the return of the Chinese elk and the harmony between people and the land	Chinese Elk Park	1. Students use AI to collect data such as temperature, humidity, and air pressure to analyze the differences in natural conditions in elk habitats.	1. ERNIE Bot analyzes the data collected by students in real time and presents the correlation between natural conditions and ecological behaviors.	Help students intuitively understand the importance of the natural environment to the survival of species, and explore the dynamic changes in the relationship between humans and land through actual data.
			2. Using AI-generated ecological data, students study the impact of human activities on the habitat of elk.	2. ERNIE Bot generates a dynamic model of human activities and habitat suitability to assist students in drawing conclusions.
	Green Ocean: Construction of Forest Ecosystem and Human–Land Relationship	Huanghai National Forest Park	1. Using AI-supported environmental measurement tools, students measure ecological data such as soil moisture in different zones and distribution of vegetation communities.	1. ERNIE Bot analyzes students’ soil moisture and vegetation distribution data in real time, and automatically generates a visual report on the relationship between environmental factors and ecosystems.	Guide students to understand the impact of human activities on forest ecology through data and behavioral records, and explore how to achieve sustainable use of forest resources.
			2. Record the impact of tourist behavior and forest management measures on ecological balance, and use AI to generate a report on the ecological impact of behavior.	2. ERNIE Bot identifies visitor behavior data and provides automatic analysis and management optimization suggestions for ecological disturbance.
	Migratory bird paradise: practice of ecological balance and sustainable development	Tiaozini Wetland	1. Students use telescopes to record the behavioral characteristics of migratory birds, and AI helps to count population data and analyze its changing trends.	1. ERNIE Bot processes the migratory bird observation data uploaded by students and generates a dynamic analysis report on population size and behavioral characteristics.	Through observation and data analysis, students can understand the dynamic relationship between human activities and the natural environment and propose ecological protection strategies for sustainable development.
			2. Use AI to simulate the impact of different development or protection measures on wetland ecosystems and propose a balance between protection and development.	2. ERNIE Bot’s ecological simulation tool generates an ecological impact prediction model based on the scenarios input by students, supporting students in optimizing strategy design.
Post-trip learning	Ecological protection plan display	Classroom	1. Student teams present wetland protection plans, including data analysis results and innovative suggestions.	1. ERNIE Bot generates group data analysis reports to quantitatively evaluate the scientific nature of student solutions.	Guide students through the complete path from data to decision-making, and cultivate innovation capabilities and awareness of sustainable development.
			2. AI evaluates the scientificity and innovation of the solution, generates feedback reports, and provides optimization directions.	2. ERNIE Bot provides improvement suggestions based on ecological theory to support students in further optimizing their solutions.
	Community Outreach and Environmental Action	Community	1. Students make environmental protection posters or videos, and AI helps to proofread the scientificity and logic of the content.	1. ERNIE Bot reviews the scientific validity of students’ promotional content and provides optimization suggestions.	Transform students’ learning outcomes into practical actions and enhance their sense of social responsibility and leadership.
			2. Use AI interactive platforms to promote wetland protection concepts, record community feedback and generate data reports.	2. ERNIE Bot analyzes community feedback data and automatically generates campaign impact assessment reports to support improvements in subsequent environmental protection actions.

.

4. Evaluation System

The evaluation system plays a vital role in the design of research and study courses. Its core goal is to comprehensively evaluate students’ ability performance in all aspects of the entire research and study process, including knowledge mastery, practical ability, innovative thinking, and the degree of understanding of the core concept of “human–land coordination” [27]. In the design of this research and study course, we adopted three core evaluation dimensions: “land to people”, “people to land”, and “human–land coordination” [35]. Each dimension reveals the interaction between human activities and the natural environment from different perspectives, and can help students deepen their understanding of the core geographical literacy of “human–land coordination” in real geographical contexts. The following is a brief description of each dimension and its important role in geographical research and study.

The impact of “land on people”: This dimension focuses on the impact of the natural environment on human activities, helping students understand how the natural environment affects human survival, development, and social activities through factors such as climate, topography, and biodiversity. During the study process, students understand the importance of wetland resources to migratory birds and human production and life through field observations and data analysis. This dimension can cultivate students’ sensitivity to the natural environment and its constraints, enhance their understanding of regional geographical characteristics, and help students understand the profound impact of nature on humans from the perspective of ecosystem functions [36].

Impact of “people on land”: This dimension focuses on how human activities affect the natural environment, especially the changes in the ecosystem caused by human activities such as agriculture, industry, and tourism development. Through field visits to migratory bird habitats, students can intuitively understand the negative impacts of human activities on wetland ecosystems (such as pollution and habitat destruction) as well as positive ecological restoration measures. This dimension cultivates students’ ability to analyze and evaluate the long-term impact of human activities on the natural environment, stimulates their ability to explore sustainable development paths, and enhances their environmental awareness and sense of responsibility [37].

“Human–land relationship coordination”: This dimension explores how to promote human economic and social development and achieve harmonious coexistence between humans and nature while maintaining ecological balance. Through data analysis and group discussions, students propose innovative solutions between wetland protection and community development, such as the development of ecotourism and wetland restoration plans. This dimension helps students comprehensively consider the complex relationship between humans and the environment, exercises their dialectical thinking and innovation ability, and cultivates students to propose coordination strategies that take into account both ecological protection and social development [38].

In order to achieve the course objectives, the evaluation system of this course is divided into two parts: process evaluation and final evaluation. Through the deep support of the intelligent tool ERNIE Bot, dynamic monitoring and accurate feedback are ensured [33]. AI technology plays a key role in this process. First, it provides an accurate evaluation basis through real-time data analysis. AI can analyze students’ data collection, task completion, and teamwork performance in research and study activities in real time; quantify students’ performance; and generate core data for process evaluation. These data can not only reflect students’ immediate performance in the course, but also provide an important basis for subsequent learning strategy adjustments [39]. Secondly, AI generates personalized learning reports based on students’ performance in the tasks. These reports not only summarize students’ strengths, but also accurately identify their shortcomings and provide practical suggestions for improvement [34]. Through these personalized reports, students can understand their learning progress more clearly and take more targeted measures to improve themselves [40]. Finally, AI also provides students with real-time interactive feedback through a dynamic feedback system [41]. This system can help students intuitively understand their learning effects and shortcomings, further optimize their learning paths, and ensure that they can achieve better growth and improvement in research and study activities [42].

4.1. Process Evaluation

Process evaluation aims to record students’ performance in research and study activities, including knowledge mastery, practical operation ability, teamwork, and depth of thinking [43]. Through this evaluation, students can reflect on their performance in the task after each activity and adjust their learning strategies in a timely manner [44].

As shown in

Table 3. Study tour process evaluation table.

Evaluation Dimensions	Evaluation Project	Evaluation Criteria	AI Analysis Support
Knowledge Mastery	Wetland ecosystem and migratory bird habitat	Whether students can accurately understand the dynamic changes in wetland ecosystems and the ecological significance of migratory bird migration.	ERNIE Bot evaluates students’ depth of understanding of ecosystem and migratory bird data based on their answers and data analysis results.
	Forest ecosystem and the impact of human activities	Can students analyze the environmental factors of forest ecology and the ecological impacts of human activities?	ERNIE Bot quantifies students’ ability to process soil moisture and vegetation distribution data and generates charts for display.
Practical Operation	Data collection and processing capabilities	Whether students can correctly use instruments (such as telescopes and soil moisture meters) to complete data collection and use AI to process data.	ERNIE Bot detects the accuracy and completeness of data collection and generates real-time feedback.
	Simulation of ecological models and predictive capabilities	Can students use AI to generate ecological simulation models and make predictive conclusions?	ERNIE Bot evaluates the scientificity of students’ models and the rationality of their predictions, and quantifies the accuracy of the simulation parameters used.
Teamwork	Task division and collaboration	Whether students have clear division of roles in group work and whether they can effectively communicate and collaborate with team members to complete tasks.	ERNIE Bot analyzes the efficiency and contribution ratio of team collaboration by recording task submission time and discussion records.
Depth of thinking	Data analysis and reflection	Can students draw scientific conclusions based on data and make suggestions for improvement based on reflection?	ERNIE Bot automatically analyzes students’ data reports, marks logical loopholes and provides directions for improvement.

, with the help of ERNIE Bot’s real-time monitoring and data analysis, the evaluation system is more accurate and efficient. During the study activities, the migratory bird data, soil moisture, and other information collected by students will be analyzed in real time by ERNIE Bot to generate a visual data report. ERNIE Bot will also track the completion of students’ tasks and quantify their performance. After each activity, ERNIE Bot will automatically generate a phased feedback report covering the advantages and improvement directions of the group and individuals, providing students with continuous growth support.

4.2. Summative Evaluation

The summative evaluation is used to comprehensively assess students’ overall performance throughout the research and study course, including knowledge mastery, skill improvement, and the understanding and application of the core literacy of ‘human–land coordination’, as well as individual and group achievements and learning attitudes [34]. The summative evaluation not only includes individual student performance, but also integrates the group performance and the students’ final project results. The summative evaluation form combines research and study reports, research and study manuals, photographic works, and other forms to ensure the comprehensiveness of the evaluation system. AI helps quantify students’ comprehensive performance in this process and generates accurate personalized learning reports [41]. During the evaluation process, ERNIE Bot will summarize and analyze students’ task data, program design and presentation, and generate personalized learning reports [42]. As shown in

Table 4. Summative evaluation.

Evaluation Dimensions	Evaluation Project	Evaluation Criteria	AI Analysis Support
Knowledge Application	Analysis and application of wetland and migratory bird data	Can students accurately explain the results of data analysis in their presentations and explain the phenomena using theories?	ERNIE Bot analyzes the accuracy and logic of the data used in the presentation and assesses the depth of students’ integration of theory and practice.
	A balanced approach to ecological protection and development	Can students design feasible plans that take into account both ecological protection and economic development, and clearly express their scientific basis?	ERNIE Bot quantifies the innovations in the solution and generates targeted improvement suggestions.
Practical results	Data analysis and visualization	Can students complete analysis based on the collected data and create dynamic maps or charts?	ERNIE Bot detects the scientificity and completeness of charts and provides visual improvement directions.
	Model design and prediction capabilities	Can students use AI to generate ecological prediction models and verify their rationality?	ERNIE Bot analyzes model parameters and evaluates the accuracy and scientificity of predictions.
Teamwork	Group task division and results integration	Whether the team can efficiently complete the division of tasks and integrate the presentation to form systematic results.	ERNIE Bot counts team members’ task participation and contribution ratios to evaluate the efficiency of team collaboration.
Innovative thinking	Ecological protection innovation	Whether students can propose innovative strategies in their plans and provide theoretical and data support.	ERNIE Bot identifies the innovative points in the solution and provides optimization directions.

, by quantifying the overall performance of the group, ERNIE Bot will also compare the innovation and scientificity of the programs of different groups [43]. In addition, ERNIE Bot combines students’ knowledge mastery, practical results, and innovative thinking to generate a comprehensive score to comprehensively evaluate students’ performance [44].

4.3. AI Evaluation Summary and Feedback

After each research activity, a discussion session will be set up to encourage students to reflect on their performance through self-evaluation and mutual evaluation, and summarize their learning gains under the guidance of teachers [45]. The content of the discussion will cover the students’ shortcomings and improvement points in practical operations, as well as their progress in thinking depth and innovation [36]. The teacher will provide written feedback after the discussion to help students further improve the research effect [35]. As shown in

Table 5. The multi-dimensional role of AI in evaluation summary and feedback.

Functional Category	Specific Content	The Role of AI Support
Generate personalized learning reports	1. Summarize students’ knowledge, practical ability and innovative program design, generate personalized learning reports, and mark problems and strengths.	ERNIE Bot points out errors in the processing of migratory bird data in the data analysis report and provides directions for improvement.
		ERNIE Bot quantifies the innovativeness of solution designs and provides more scientific optimization suggestions.
Real-time dynamic feedback	1. In the process of evaluation, students can view the task completion progress and analysis results in real time.	Through ERNIE Bot’s interactive feedback system, students can intuitively understand their performance and shortcomings in each aspect.
	2. In summative evaluation, generate suggestions in real time based on the presentation content to support student learning improvement.	ERNIE Bot provides dynamic progress and improvement suggestions for task completion to enhance the targeted nature of learning.
Improved learning paths	1. Based on students’ comprehensive performance, recommend personalized learning improvement plans to help students make up for their shortcomings.	ERNIE Bot recommends more detailed data analysis guidance content to students with weaker performance.
	2. Provide more challenging tasks for high-performing students.	ERNIE Bot provides high-performing students with more complex ecological protection problem tasks and cultivates advanced analytical and problem-solving skills.

, in the entire evaluation process, AI not only undertakes the tasks of data analysis and report generation, but also helps students and teachers optimize the learning process through real-time feedback and personalized suggestions.

The evaluation system is deeply empowered by AI technology, which not only improves the objectivity and efficiency of evaluation, but also realizes personalized and dynamic feedback support. Under the guidance of AI, students can understand their own learning strengths and weaknesses and optimize their learning paths; teachers can adjust their teaching strategies through accurate evaluation data, and ultimately improve the effectiveness and quality of the entire research course.

4.4. Test and Effect Evaluation

4.4.1. Process Test

The process test aims to comprehensively and dynamically monitor the development of students’ abilities and knowledge mastery during the course study through a variety of test forms and high-frequency assessments, discover students’ learning strengths and weaknesses in a timely manner, and provide a basis for personalized learning support and teaching strategy adjustments. The specific test content is shown in

Table 6. Process test content.

Test Category	Specific Test Content	Test Format	Scoring Subject and Method
Knowledge mastery test	Knowledge on wetland ecosystem and migratory bird habitat	Weekly online quiz (5 multiple-choice questions, 3 short-answer questions, 2 analytical questions)	ERNIE Bot is evaluated based on answering questions and data analysis results, for teachers to refer to
	Knowledge on forest ecosystems and the impact of human activities	Group discussion and report every two weeks	Teachers grade students based on discussion performance (participation, accuracy, and depth of ideas) and presentation content
Practical skills test	Data collection and processing capabilities practice	After the field investigation, data will be collated and analyzed and a report will be submitted	ERNIE Bot is scored based on data accuracy, completeness, and analysis rationality
	Simulation ecological model and predictive capability practice	Use tools to build models and predict ecological change trends	Teachers and ERNIE Bot jointly evaluate the scientific nature of the model, the rationality of the prediction, and the ability to apply knowledge tools
Teamwork ability test	Task division and collaboration observation	During each group activity, the teacher observes and ERNIE Bot records	Teachers make observations and ERNIE Bot analyzes team collaboration efficiency and other scores based on records.
	Teamwork project assessment	Monthly teamwork project presentation and evaluation (with the participation of other groups and teachers)	Other groups, teachers, and ERNIE Bot will rate based on project goals, division of labor, collaboration, results, etc.
Depth of thinking test	Data analysis and reflection report	Write a report after completing data collection and analysis	ERNIE Bot analysis report logic, depth and reflection effectiveness rating
	Problem-solving and Creative thinking challenges	Set questions from time to time and discuss solutions in groups	Teachers and ERNIE Bot evaluate the innovation, feasibility, flexibility, and depth of thinking of the solution

.

4.4.2. Summative Test

The summative test focuses on the comprehensive evaluation of the comprehensive learning outcomes achieved by students after the completion of the entire research course, and considers students’ mastery of course content, ability improvement level, and understanding and practical ability in terms of core literacy from multiple dimensions, such as knowledge application, practical results, and overall team performance, to determine whether students have achieved the expected goals of the course. The specific content is shown in

Table 7. Summative test content.

Test Category	Specific Test Content	Test Format	Scoring Subject and Method
Comprehensive test on knowledge application	Research results report and defense	After the course, the group will present and defend their research	Teachers grade the reports based on accuracy, logic, depth, and the degree of “coordination between humans and land”. ERNIE Bot helps analyze the data.
	Study manual and knowledge summary assessment	Check the study manual and personal knowledge summary	Teachers grade the construction and deepening of the knowledge system based on the comprehensive accuracy, depth of knowledge understanding, and degree of integration with practice in the manual. ERNIE Bot assists in data analysis.
Comprehensive evaluation of practical results	Field investigation results display and evaluation	Present field investigation materials and analyze and propose solutions	Teachers grade the data based on the quality of the data, scientific analysis, and feasibility of the solution, and ERNIE Bot helps analyze the data
	Evaluation of innovative practice works	Submit innovative practice works (posters, videos, plans, etc.)	Teachers evaluate from multiple dimensions such as innovation, scientificity, practicality, and artistry, and ERNIE Bot helps analyze the rationality of data application, etc.
Overall team performance evaluation	Quality assessment of team project results	Comprehensive evaluation of team projects (publicity activities, protection program implementation, etc.)	Teachers conduct field visits, community feedback surveys, and project report scoring, and ERNIE Bot assists in data analysis
	Team member growth and development analysis	Compare the performance of team members before and after the course, and analyze the role of collaboration in individual growth	Teachers conduct comprehensive evaluation based on observation and students’ self-evaluation and peer evaluation, and ERNIE Bot assists in data analysis

.

4.4.3. Test Results Before and After the Course

1. Knowledge mastery. In the test of wetland ecosystems and migratory bird habitats, at the beginning of the course, the average score of students was about 60 points, the correct rate of multiple-choice questions was about 60%, the answers to short-answer questions were relatively brief, and the analytical questions generally lacked depth. As the course progressed, after multiple knowledge tests and learning activities, the average score in the middle of the course increased to about 75 points, the correct rate of multiple-choice questions increased to 80%, the short-answer questions were able to explain the principles in more detail, and the analytical questions began to be analyzed in depth with actual cases. At the end of the course, the average score reached more than 85 points, the correct rate of multiple-choice questions stabilized at more than 90%, the answers to short-answer questions were accurate and in-depth, and the analytical questions were able to comprehensively and deeply analyze the complex relationship between wetland ecological changes and migratory bird habitat changes. In the group discussion and report on the knowledge of forest ecosystems and the impact of human activities, at the beginning of the course, students had low participation, superficial views, and an insufficient ability to integrate and apply knowledge, with an average score of about 65 points. After a series of learning and practical activities, the students’ participation in the course was significantly improved in the middle of the course. They were able to put forward some valuable ideas, and their ability to integrate and apply knowledge was enhanced. The average score increased to about 80 points. At the end of the course, the students were active in the discussion, able to deeply analyze the relationship between forest ecosystems and human activities, and could use the knowledge they learned to put forward reasonable protection and development suggestions, with an average score of more than 90 points.

2. Practical operation ability. In the practice of data collection and processing ability, at the beginning of the course, students had major problems in the accuracy of data collection, such as large errors in the statistical number of migratory birds and inaccurate soil moisture measurements. Data processing was also relatively unfamiliar, and the average score was only about 50 points. With the increase in the number of practices and the real-time guidance of ERNIE Bot, the accuracy of the students’ data collection increased to more than 80% in the middle of the course, and they were able to skillfully use tools for data processing, with an average score of about 75 points. At the end of the course, the accuracy of the students’ data collection reached more than 95%, and they were able to efficiently and accurately complete data processing and analysis, with an average score of more than 90 points. In the practice of simulating ecological models and predictive ability, at the beginning of the course, the models constructed by students had many scientific problems, the prediction results were unreasonable, and the average score was about 45 points. With the deepening of learning and the proficiency of the ERNIE Bot simulation tool, the scientificity and predictive rationality of the model in the middle of the course were significantly improved, and the average score increased to about 70 points. At the end of the course, students were able to build a more complex and scientific ecological model, and the prediction results were highly reliable, with an average score of more than 85 points.

3. Teamwork ability. In the task division and collaboration observation, at the beginning of the course, some teams had problems such as unreasonable task allocation and poor communication among members, and the team collaboration efficiency was low, with an average score of about 60 points. After the guidance of teachers and the running-in of multiple team activities, the team collaboration situation in the middle of the course was significantly improved, the task division was more reasonable, the communication was smoother, and the average score increased to about 80 points. At the end of the course, the team members cooperated well and were able to complete the tasks efficiently, with an average score of more than 90 points. In the evaluation of teamwork projects, at the beginning of the course, the group projects had problems such as unclear goals and insufficient innovation in results, with an average score of about 65 points. With the improvement of teamwork ability and the cultivation of innovative thinking, the quality of the mid-course projects improved significantly, with clear goals and certain innovation in results, and the average score increased to about 85 points. At the end of the course, the group projects performed well in terms of goal achievement, innovative design, and results presentation, with an average score of more than 95 points.

4. Depth of thinking. In the data analysis and reflection report, at the beginning of the course, the logic of the student report content was poor, the reflection was not in-depth, and the improvement measures were not targeted, with an average score of about 55 points. As the course progressed, students gradually learned to analyze data in depth, their logical thinking ability was enhanced, their reflection was more profound, and their improvement measures were operational. The average score in the middle of the course increased to about 75 points. At the end of the course, students were able to write high-quality data analysis and reflection reports, with a high level of thinking depth and logic, and an average score of more than 90 points. In the problem solving and innovative thinking challenges, at the beginning of the course, the solutions proposed by students were relatively conventional, lacked innovation, and their thinking was not flexible enough, with an average score of about 50 points. After many challenges and much learning, students began to propose some novel solutions in the middle of the course, their thinking flexibility improved, and the average score increased to about 70 points. At the end of the course, students were able to propose innovative and feasible solutions to complex geographical problems, and their thinking depth and innovation ability were significantly improved, with an average score of more than 85 points.

5. Comprehensive test of knowledge application. In the report and defense of research results, the average score after the end of the course reached 88 points. In their reports, students were able to accurately apply geographical knowledge, deeply analyze the relationship between the Yellow Sea (Bohai Sea) migratory bird habitat ecosystem and human activities, and clearly explain the application of the “human–land coordination concept” in actual cases. For example, when explaining the relationship between wetland protection and the economic development of surrounding communities, they were able to use geographical principles to propose reasonable coordination plans. During the defense of their research, students were able to accurately answer questions raised by experts, showing a deep understanding and proficient application of knowledge. In the evaluation of the study manual and knowledge summary, the average score was 90 points. The students’ study manuals were detailed and accurate, not only containing the key points of geographical knowledge, but also recording the observations and thoughts during the field investigation. The personal knowledge summary can systematically sort out the knowledge learned during the study, deeply understand the connection between geographical concepts and principles, and reflect the construction and deepening of the knowledge system.

6. Comprehensive evaluation of practical results. In the display and evaluation of field investigation results, the average score was 92 points. The field investigation materials presented by the students are rich and detailed, and can accurately reflect the current situation and problems of wetland and forest ecosystems. The analysis of the data is in-depth and scientific. For example, by comparing photos of migratory bird habitats in different periods, the causes of ecological changes are accurately analyzed, and targeted protection measures are proposed. The solutions are highly feasible. For example, the proposed wetland ecological restoration plan combines local actual conditions and takes into account multiple factors such as ecology, economy, and society. In the evaluation of innovative practice works, the average score was 90 points. The students’ innovative practice works were outstanding in terms of innovation. For example, some propaganda posters used unique perspectives and creative techniques to attract the audience’s attention to ecological protection issues; the ecological protection plan proposed a novel ecotourism model that takes into account ecological protection and economic development. The scientificity and practicality of the works have also been recognized by experts. For example, the construction of the ecological prediction model conformed to scientific principles and can provide reference for actual decision-making.

7. Evaluation of the overall performance of the team. In the quality evaluation of the team project results, the average score was 95 points. In the ecological protection community publicity activities, the team effectively raised the ecological protection awareness of community residents through various forms (such as holding lectures, distributing publicity materials, organizing field experiences, etc.), and the participation in the activities was high, which was highly recognized by the community. In the implementation of the migratory bird habitat protection plan, the team was able to formulate detailed plans based on actual conditions and effectively implement them, achieving significant ecological protection results, such as an increase in the number of migratory birds and improvements in the habitat’s ecological environment. In the analysis of the growth and development of team members, it was found that team members had significantly improved in terms of knowledge mastery, practical operations, and thinking patterns. For example, students who originally had a weak foundation in geography knowledge were able to skillfully use knowledge to solve practical problems through team collaboration and learning; the collaboration and communication skills among members were greatly exercised, forming a good team atmosphere, and promoting the all-round development of individuals.

4.4.4. Student Feedback on AI-Enabled Teaching Methods

1. Improved learning interest. More than 90% of the students said that the application of AI technology makes the learning process more vivid and interesting. For example, the 3D wetland model generated by ERNIE Bot made them feel as if they were there, enhancing their intuitive feeling of the geographical environment and thus stimulating their interest in learning. Students generally believe that the personalized learning path provided by AI can meet their different needs, allowing them to choose learning content according to their interests. For example, students interested in migratory bird ecology could explore related knowledge in depth, which improved their autonomy and enthusiasm for learning.

2. Improved learning efficiency. About 85% of the students reported that the powerful functions of AI tools for data processing and analysis have greatly saved time. When processing a large amount of migratory bird observation data and wetland ecological monitoring data, ERNIE Bot can quickly and accurately complete the analysis, helping them to obtain valuable information in a timely manner, thereby improving learning efficiency. The real-time feedback mechanism is highly praised by students. In total, 75% of the students said that during the practical operation, ERNIE Bot could immediately point out problems and provide improvement suggestions, enabling them to adjust their learning strategies in time and avoiding wasting time in the wrong direction.

3. Improved perception of ability. In total, 95% of the students believe that their practical operation ability has been significantly improved through AI-enabled research courses. When using AI-assisted instruments for data collection and processing, they not only mastered advanced technical means, but also learned how to combine data with actual geographical phenomena for analysis, improving their practical hands-on ability. About 90% of the students feel that their innovative thinking ability has been exercised. With the support of virtual research scenarios and ecological simulation tools provided by AI, they were able to try different solutions and cultivate innovative awareness and innovation ability.

4. Overall evaluation of teaching methods. More than 95% of students are satisfied or very satisfied with the AI-enabled geographical research travel teaching method. They believe that this teaching method breaks the limitations of traditional teaching, allowing them to better understand and apply knowledge in real geographical situations, while cultivating a variety of abilities, which has a positive role in promoting their learning and growth.

5. Limitations of and Optimization Strategies for AI-Driven High School Geography Research Design

The application of AI in high school geography research and study provides great potential for the innovation of teaching models. However, in practice, AI-driven geography research and study design also faces some limitations that cannot be ignored. These limitations involve both technical bottlenecks and difficulties in teaching implementation and resource allocation. Therefore, exploring these limitations and proposing optimization strategies are of great significance to further improving the application effect of AI in geography research and study.

5.1. Limitations of AI in High School Geography Research

• Insufficient technical adaptability

AI technology can achieve data analysis, dynamic feedback, and personalized support in geographical research and study, but its effectiveness is highly dependent on high-quality data input and the support of technical platforms. In ecological monitoring and geographic information system (GIS) applications, real-time updated data are particularly needed. However, in some remote areas, environmental data may be missing or inaccurate [46]. This problem is particularly prominent in the context of rural revitalization. Rural primary schools in some areas may face the problem of incomplete data collection, which affects the application effect of AI systems. For example, in the study activities of migratory bird habitats in the Yellow (Bohai) Sea, wetland ecological monitoring data may not cover all key nodes, which directly leads to certain deviations in the ecological model predictions generated by AI [47]. In this way, although AI can process and analyze large amounts of data, if there are loopholes or errors in the basic data, the final analysis results may be inaccurate, thus affecting students’ understanding of and practice in the ecosystem.

In addition, the use of AI technology is highly dependent on the internet and high-performance devices [35]. In research and study activities, students need to access and operate AI through internet connections and advanced devices, but in some cases where the network is unstable or the equipment is in poor condition, all functions in the research and study design may not be realized. This technical dependence limits the widespread application of AI in different regions and schools, especially in some resource-scarce regions, where the advantages of AI technology in education may not be fully utilized.

2.

Differences in students’ abilities and learning habits

Although AI technology can support personalized learning, its application effect is still affected by students’ basic abilities and learning habits [48]. Some students may lack the necessary data analysis skills or have insufficient understanding of AI, which may limit their participation and learning effects in research and study activities [3]. In particular, when operating AI, students of different abilities may face different difficulties [49]. For example, in the process of ecological model design and data visualization, some students may not fully understand the generated results, which affects their in-depth understanding and practical application of the learning content [50].

In addition, some students are overly dependent on AI-generated feedback, which may weaken their ability to think independently and practice [51]. Although AI can provide rapid analysis and feedback, over-reliance on AI may lead to students’ lack of independent problem-solving ability, which in turn affects their critical thinking and innovation ability [52].

3.

Transformation and adaptation of teachers’ roles

The introduction of AI technology has placed higher demands on the role of teachers. Teachers not only need to master traditional teaching skills, but also need to be familiar with the operation and application of AI technology and be able to effectively guide students to use AI to complete tasks [53]. However, some teachers may face challenges in adapting and applying technology, especially in highly technical aspects such as data analysis and model design [54]. The existence of technical barriers means that the depth of teachers’ understanding of AI directly affects the effectiveness of teaching design [50]. If technical training is insufficient, teachers may find it difficult to fully utilize the functions of AI, which will affect students’ learning experience and the overall effect of research and study activities.

At the same time, the deep involvement of AI has also posed new challenges to the role of teachers. In traditional teaching, teachers are often in a dominant position, but the introduction of AI may weaken this dominance and give more control over learning to students and technology [44]. In this context, finding a balance between technical support and teaching guidance, taking advantage of AI while retaining the leading role of teachers, has become an important issue. This requires teachers to not only be users of AI, but also become a bridge between students and technology, redefining their roles and responsibilities in the context of AI-assisted teaching [10].

5.2. Optimization Strategy

In view of the above limitations, we propose the following optimization strategies to further improve the application effect of AI in geographical research design.

• Improve data quality and technology platforms

To ensure the integrity and real-time nature of the data, a regional ecological data sharing platform can be established to cooperate with scientific research institutions and government departments. For example, cooperation with the Yellow (Bohai) Sea Bird Conservation Organization to obtain more accurate migratory bird data. This will not only help improve the accuracy of the AI model, but also promote the coordinated development of ecological protection and scientific research. At the same time, in order to support the smooth progress of research and study activities, it is necessary to provide a stable internet environment and high-performance equipment, especially in remote areas. Considering the special conditions in these areas, the application of portable devices and offline functions is particularly important, can effectively ensure the smooth development of research and study activities, and can continue to collect and analyze data even in an unstable network environment.

2.

Strengthen students’ technical training

During the pre-trip thinking stage, special training on the use of AI should be introduced to ensure that students master basic data processing and model operation skills. This will help students use AI more confidently in research and study activities, thereby improving learning efficiency and task completion. At the same time, in order to adapt to the differences in the abilities of different students, tiered teaching can be designed to provide basic and challenging task options. In this way, students can choose appropriate tasks to participate in according to their ability levels, which can ensure that students with weaker basic abilities can also obtain an effective learning experience, and sufficient challenges for students with higher abilities can be provided, stimulating their innovative thinking and interest in in-depth exploration. This tiered design helps to enhance each student’s sense of participation and accomplishment.

3.

Improve teachers’ AI application capabilities

In order to improve teachers’ AI application capabilities, systematic AI technology application training can be provided to teachers, covering GIS operations, ecological model construction, dynamic data analysis, and other aspects. This will help teachers better master the use of AI and improve their technical application capabilities in teaching. At the same time, in order to reduce the technical burden on teachers, an AI teaching assistant system can be introduced to assist teachers in task assignment, data analysis, and feedback report generation. With the support of AI teaching assistants, teachers can focus more on teaching design and student guidance, improve teaching efficiency and effectiveness, and ensure the smooth application of AI technology in the classroom.

4.

Enrich AI functions and teaching scenarios

In order to stimulate students’ interest in learning, personalized learning resource recommendation options can be provided to students. For example, relevant case studies and the latest scientific research results in the field of ecological protection can be provided so that students can be exposed to richer and more cutting-edge knowledge. This can not only enhance students’ interest in the subject, but can also help them better understand practical applications and scientific research trends.

At the same time, AI technology can be combined with other disciplines (such as biology, environmental science, etc.) to design interdisciplinary tasks. This interdisciplinary integration approach helps to improve students’ comprehensive literacy and cultivate their systematic thinking and ability to solve complex problems. By combining knowledge and skills from different fields, students can not only master AI technology, but can also have a deeper understanding of the connection and application between disciplines, thereby improving their comprehensive learning ability and innovation ability.

5.3. Application Prospects After Optimization

Through the above optimization strategies, the application of AI in geographical research will be more efficient and comprehensive. On the basis of mastering AI, students can complete tasks more confidently and propose scientific and innovative solutions. Personalized learning resource recommendations and hierarchical task design will greatly enhance students’ sense of participation and achievement, and inspire their active exploration spirit in ecological protection and geographical research. At the same time, the role of teachers has transformed from a simple knowledge transmitter to a guide for the use of AI and a feedback analyst of the learning process. With the help of AI, teachers can perform data analysis and student performance evaluation more efficiently, provide more targeted guidance and support, and improve teaching effectiveness. Based on high-quality data and optimized AI functions, geographical research courses can be expanded to more ecological scenarios, cover a wider range of subject knowledge, dynamically update course content, and cultivate students’ more comprehensive subject literacy and sustainable development awareness. This innovative course model supported by AI not only improves learning efficiency and teaching quality, but also promotes the transformation of the education model and promotes students to master the ability to solve real problems in practice.

6. Conclusions

With the continuous development of AI technology in the field of education, we focused on its exploration in the design and application of high school geography study and research. Taking the China Yellow (Bohai) Sea migratory bird habitat study and research course as an example, we have achieved a series of significant results.

First, in terms of design concept, we used AI (ERNIE Bot) as a support, integrated the problem situation setting, the learning resource optimization and the dynamic feedback mechanism, and formed an innovative research and study course design concept. By providing dynamic data support, personalized learning paths, immersive research and study experiences, and a diversified evaluation mechanism, it significantly improves students’ learning experience and ability training.

Secondly, in terms of course content, we focused on the “coordination of people and land”, covering pre-trip thinking, on-trip research, and post-trip learning. In each link, with the help of a variety of practical activities and tools, such as field investigations, data collection, ecological monitoring, maps, remote sensing images, and data analysis tools, AI technology is deeply integrated to optimize students’ learning paths and task completion results.

Subsequently, we established an evaluation system based on the three dimensions of “land to people”, “people to land” and the “coordination of human–land relationships” to comprehensively evaluate students’ performance in terms of knowledge mastery, practical ability, innovative thinking, and understanding of the “coordination between people and land”.

With the help of ERNIE Bot, process evaluation and final evaluation are intelligentized. In the process evaluation, the students’ data collection, task completion, and teamwork performance were monitored in real time, and visual data reports and phased feedback reports were generated; in the final evaluation, the task data, program design, and presentation were summarized and analyzed, and personalized learning reports and comprehensive scores were generated. This not only improves the objectivity and efficiency of evaluation, but also provides personalized and dynamic feedback support to students and teachers, effectively optimizing learning and teaching strategies.

In general, we have provided a brand-new teaching model for high school geography education, organically combining traditional geographical research with modern AI technology, breaking the limitations of traditional teaching, and providing practical reference for the innovative development of geography teaching.