| 研究実績の概要 |
The researcher started his project in October by targeting analysis of land cover changes in agricultural landscapes. Since October, the researcher has produced seven national rice fields maps for Japan since the 1980s. This will allow the continuation of the project by assessing the causes of farmland birds shifts on distributions and populations.
The rice maps were produced using all available Landsat satellite images from the 1980s. Landsat images were pre-processed and temporally aggregated based on cloud cover and image availability. By combining field-based data of temperature and rice field transplantation dates, a phenology algorithm was built to classify the satellite images and to obtain national scale rice maps for seven time periods. These processes were performed using Google Earth Engine and python. The accuracy of the maps was assessed by using the JAXA 2014-16 Land Cover Map and using visual interpretation of aerial photography. Changes in rice fields area were analyzed at a national and prefectural levels. A decrease of almost 28% of the rice fields was recorded since the 1985 in Japan based on our maps.
A web application was created to facilitate the exploration of the rice maps and the rice field area changes at national and prefectural levels. The web application was developed using R Shiny and it will be accessible online soon. This web application will be useful for researchers, managers and policy makers concerned about agricultural changes in Japan and its potential impacts in biodiversity and people.
|
| 現在までの達成度 (区分) |
現在までの達成度 (区分)
2: おおむね順調に進展している
理由
Despite the changes in the order of the research packages within the research proposal, the researcher has almost completed, within six months, most part of Objective 2.1 of the proposal: farmland mapping at a national scale. This is the foundation of the rest of the project, so the researcher has made sure to dedicate enough time to create a high-quality product. This will potentially allow the rest of the project to succeed and will open new lines of research within the proposal topic.
The producing of the agricultural maps was done faster than expected. However, because some pre-analysis needed to start the next phase of the project were not yet completed, we cannot consider this as a 100% done. In any case, the project team considered that the completion of the rice maps was fundamental to the rest of the project and therefore prioritized it. We consider that this will be beneficial for the completion of the project as a whole.
In addition, an extra product, not considered in the proposal, was produced. A web application were researchers and other public can visualize our maps was built and will be publicly available soon. This covers parts of the proposal in terms of publishing our research results and doing outreach to the general public, which was planned for the second year of the fellowship.
|
| 今後の研究の推進方策 |
The plans for the research scheme for this year are divided in three stages. First, a manuscript describing our rice field maps and their production methodology will be completed and sent for publication in a scientific journal. We expect that this will be published by the end of 2021.
The second stage will consist of the analysis of the causes of rice fields decline in Japan since the 80s. Understanding the causes of rice fields declines in Japan is fundamental to protect agricultural landscapes and to predict areas with potential risk of rice field abandonment. To do this, the researcher will use the produced rice field maps and identify the areas were rice fields disappeared. This will be separated into abandoned rice fields and fields converted to developed land. After that, a spatially explicit modelled will be applied to identify environmental and socio-economic factors that explain rice field decreases. This analysis will be multi scale and it will consider spatial correlation.
The third stage will consist of studying the relationships between rice field changes and the distribution and population changes of farmland birds in Japan. To do that, we will use the Japanese Breeding Bird Atlas. We will subset farmland species and obtain population and distribution changes at the national and prefectural levels. We then will use data on farmland changes to explain trends in distributions and populations. We hope that this analysis will shed some light on the high variability of the trends for different farmland species in different regions of Japan.
|
