why I can not encode and decode Chinese characters in iso-8859-1 encoding format but I can convert it just fine without any problems on utf8 format

...ahem, but because utf8 covers 1,112,064 codepoints, while iso-8859-1 only 255 and can barely handle European repertoire? You could make Perl let you know you are trying a dubious thing:

use utf8;
use Encode;
encode( 'ISO-8859-1' , '這是一個測試', Encode::FB_CROAK );

...

"\x{9019}" does not map to iso-8859-1 at ...