Keywords: Land-use changes, Climate Change, Remote sensing, and GIS,
What is the use of a fine house if you haven’t got a tolerable planet to put it on.
Henry David Thoreau,
Climate change and Land use changes are interlinked to each other. A variation on one ultimately affects the other on various spatiotemporal grounds. Nevertheless, inappropriate land uses and urban planning are the fundamental cause of Climate Change. Land-use changes and Climate Change are the global issues rising in the developed as well as developing countries. Land use is affecting the climate by pouring high concentrations of greenhouse gases into the environment, conversion of vegetative lands to provide urban infrastructures to facilitate the exponentially rising human population, and deforestation. While the outcomes of climate change resulted in the form of unpredictable precipitation, rising temperatures, and increased meteorological hazards. Hence, it is necessary to study the Land Use Land Cover (LULC) changes to determine the climate change pattern of any region.
The less we do to address climate change now, the more regulations we will have in the future.
Bill Nye,
Remote sensing and GIS techniques have proved to be useful tools for such studies. GIS specialists use this technique to study climate change patterns, Land use classifications, Carbon Management/Stocks, Sustainable Development and Planning, and Disaster Management. Remote Sensing & GIS allows specialists to investigate the Land use changes in a short period, at an inexpensive cost, with more and more accuracy and precision. Land features or land cover are easily classified by using remote sensing techniques with the help of multispectral satellite imageries.
The RS & GIS-based Land use classification techniques are of two types:
- Supervised Classification Method
- Unsupervised Classification Method
Maximum likelihood classification in ArcMap is an example of Supervised classification. However, this method is somehow criticized due to normal distribution.
Example of unsupervised classification includes: Random Forest (RF), Artificial Neural Network (ANN), Decision Tree, Fuzzyset CTA Algorithm, Fuzzy Artmap, KNN, Object Base Classification, SVM, and Expert Systems. These classification methods will be discussed later in future blogs.