Progress 05/01/24 to 04/30/25
Outputs
Target Audience:The target audiences for this project include: 1) Academic researchers a) Additional knowledge for researchers working in the field of remote sensing to detect and monitor plant populations b) Additional knowledge for researchers using deep learning and machine learning of satellite imagery to provide land use classification data products c) Additional knowledge for researchers in the field of predictive species distribution modeling, especially for invasive species with highly biased or undersampled occurrence records and/or actively expanding ranges. 2) University extension professionals a) Aid in developing materials for research and outreach on invasive species of the rangelands of the Upper Midwest b) Aid in developing materials for research and outreach on predicting invasive species distributions and developing risk assessments. 3) Natural resources land managers a) Land managers with needs for detecting and monitoring invasive species populations - specifically leafy spurge. b) Developing and understanding risk assessment for potential for future invasive species spread c) Determine the effectiveness of past eradication efforts. Changes/Problems:We have not expended funds quite as quickly as planned due to delays in hiring of qualified scientists/postdocs and researchers for the project in the first year and a half. We now have someone in place that is making good progress on goals and we were able to produce a high quality manuscript. We have found that although there was adequate transference across landscapes, some of the regional models performed better. We are using this information to help design better deep-learning detection models and develop a better strategy this season to collect more structured occurrence data for model development and validation. In addition, the collection of some of the data on biocontrol is taking longer than anticipated. Data is distributed diffusely across various digital and non-digital records at institutions at varying levels of government, especially between different states and agencies/control units. We have been making more progress within the past month to help collate the data and started gathering more contact information to obtain additional data across agencies/states. What opportunities for training and professional development has the project provided?In the past year, this grant has supported work by faculty, a graduate student, and an undergraduate researcher. Our senior researcher also attended the Upper Midwest Invasive Species Conference in November to share research results and build collaborative networks for the project across the region. How have the results been disseminated to communities of interest?We have prepared and submitted a manuscript on the results of our deep-learning and process-based SDMs. The deep-learning model detected leafy spurge across two decades. We used that information to generate a reduced-bias occurrence record. With that we were able to develop range expansion forecasts for leafy spurge. We presented the results at the Upper Midwest Invasive Species Conference in November 2024. Land managers and agricultural professionals were enthusiastic about the results. What do you plan to do during the next reporting period to accomplish the goals?We are gathering additional occurrence data using a systematic and structured sampling protocol across the Northern Great Plains to provide more testing, training, and validation data to support larger-scale models. We are testing new model architectures and leveraging additional sources of satellite imagery to build light-weight, efficient, and accurate deep learning detection models. We are continuing to reach out to partners in the natural resources community who we have been working with on this project to gather datasets on the outcomes of biological control projects, which we have begun the process of compiling, in order to further our work on the last objective of our project.
Impacts
What was accomplished under these goals?
We have made additional progress on the goals outlined in objectives 1 and 2 over the past year. We refined our previous methods and built a new temporal-CNN deep learning model using Landsat imagery with a specific focus on Minnesota. This allowed us to fine-tune the model, collect demographic information on leafy spurge populations, and interrogate the accuracy, sensitivity, and precision of the model outputs and more rigorously validate the model. We were able to determine that leafy spurge aerial coverage increased by 570% over the past two decades - from 100,670 ha to 715,600 ha. We found that the increase was highly associated with two distinct areas that were in and around urban centers in the state and invasion pathways were highly associated with roadways. We are now working on leveraging what we have learned from our MN deep-learning models to build accurate models across the Northern Great Plains. We are also continuing to develop a landscape scale validation dataset and collect historical information on biocontrol treatment status on select populations in select regions for further study.
Publications
|
Progress 05/01/23 to 04/30/24
Outputs
Target Audience:Our work is of interest to professional land managers and extension agents as well as researchers also working in the field of remote sensing. In the past year of out reseach project, we have reached out to our partnership stakeholders who support our project. We have also reached out to others within the University community - including researchers and extension professors through meetings andseminars. Changes/Problems:
Nothing Reported
What opportunities for training and professional development has the project provided?In the past year, this grant has supported work by faculty, a staff scientist, a graduate student, and an undergraduate researcher. How have the results been disseminated to communities of interest?Preliminary results for the new deep learning model for leafy spurge detection and preliminary demographic data were shared in a public dissertation defense seminar for the supported graduate student. We are currently preparing a manuscript for publication about the deep learning model and demographic data for leafy spurge in Minnesota. What do you plan to do during the next reporting period to accomplish the goals?We are planning our upcoming field season to generate additional testing, training, and validation data. We are also continuing to explore new sources of imagery and deep learning architectures to build more efficient and accurate deep learning models. We will also continue to reach out to partners in the natural resources community who we have been working with on this project to gather datasets on the outcomes of biological control projects in order to further our work on the last objective of our project.
Impacts
What was accomplished under these goals?
We have made substantial progress on the goals outlined in objectives 1 and 2 over the past year. In previous work, we developed a deep learning model to detect leafy spurge based on a complex landscape in Minnesota. We used data from North Dakota to assess the transferability of this previously built model and determined that it was able to detect leafy spurge but would require some tweaking to be fully implementable in the management space. We also built a new temporal-CNN deep learning model using Landsat imagery from across the Northern Great Plains. The preliminary model for the region is promising and we are continuing to refine the model and explore other architectures that may provide greater model sensitivity and precision for leafy spurge detection, especially in more complex habitat types. We also built a Landsat based model to detect leafy spurge across Minnesota. The state-based model performed well and we used our model to generate predictions about population growth in Minnesota. We are using this preliminary growth data to start developing model frameworks for demographically-based SDMs. We are in the process of planning the field season to collect more validation data from additional regions of the Northern Great Plains to assess the true accuracy of our models. We are also currently preparing a manuscript to publish our new deep learning leafy spurge detection model for Minnesota.
Publications
|
|