Efficient management and processing of spatial data, especially in the context of map algebra, have become crucial for researchers. The challenge lies in the effectiveness of map algebra operations on voluminous series of raster data, which are essential for analyzing and extracting valuable insights. Many compact data structures have been developed to support such operations; however, they all suffer from data representation or performance issues. This work presents the Compressed Line Raster (CL-raster), a new compact data structure that performs a line-based compression of raster data. CL-raster stores raster data in a compressed format ready for processing in a line-by-line fashion, allowing fast processing of a series of raster data with reduced memory consumption. Experiments show that our approach is efficient, outperforming the state-of-the-art competitor in terms of processing time and memory requirements up to several times.