Policy makers in Canada have suggested that the scientific community s
hould develop an indicator of biodiversity change that can be implemen
ted quite quickly without a major new investment in monitoring systems
. We propose that such an indicator can be developed from the theory o
f species gradients in community ecology. The term;species gradients'
refers to the increasing diversity of species through time under stabl
e conditions, and the increasing diversity of species with the increas
ed use of available resources. This theory is reviewed under four diff
erent headings: evolutionary ecology, the energy theory, the resource
productivity theory and the thermodynamic mechanism. The theoretical a
rguments provide a basis to propose detection of the 'leaky ecosystem'
as an indicator of biodiversity. We propose that it is possible to de
tect the leaky energetics of ecosystems by means of routinely availabl
e observations of outgoing longwave radiation.