Back

Characterizing changes in land cover and forest fragmentation in Dhorpatan Hunting Reserve of Nepal from multi-temporal Landsat observations (1993-2018)

Sharma, S.; Bista, M.; Mingshi, L.

2019-11-18 ecology
10.1101/846741 bioRxiv
Show abstract

Recent centuries have experienced drastic changes in land cover around the world where Himalayan countries like Nepal have undergone changes in the past several decades because of increasing anthropogenic pressure, natural risks and climatic factors. Accordingly, forest fragmentation has also been increasing alarmingly, which is a matter of concern for natural resource management agencies and biodiversity conservation communities. In this study, we assessed land cover change and forest fragmentation trends in Dhorpatan Hunting Reserve of Nepal by implementing landscape fragmentation and recovery process models, and calculating landscape indices based on five-date land cover maps derived from Landsat satellite images from 1993 to 2018. Six land cover types including forest, grass land, barren land, agriculture & built-up, water bodies and snow & glaciers were determined after an intensive field survey. Diverse derived image features were fed to the Support Vector Machines classifier to create land cover maps, followed by a validation procedure using field samples and reference data. Land cover maps showed an increase in forest area from 37.32% (1993) to 39.26% (2018) and snow & glaciers from 1.72% (1993) to 2.15% (2018) while a decrease in grassland area from 38.78% (1993) to 36.41% (2018) and agriculture & built-up area from 2.39% (1993) to 1.80% (2018). Barren land and water body showed negligible changes. The spatial explicit process of forest fragmentation indicated that shrinkage was the most responsible factor of forest loss while expansion was dominant to increment for forest restoration. High dependency of people persists on the reserve for subsistence resources being a cause of forest fragmentation and posing threats to biodiversity. Focus should be made on strategies to decrease the anthropogenic pressure on the reserve. This requires approaches that provide sustainable alternative resources to the local people and innovations that will help them become less reliant on natural resources.

Matching journals

The top 1 journal accounts for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.