Geo-neutrino detection can determine the amount of long-lived radioactive elements within our planet, thus providing a direct test of the Bulk Silicate Earth (BSE) model and fixing the radiogenic contribution to the terrestrial heat. We present a prediction for the geo-neutrino signal at KamLAND as a function of the Uranium mass in the Earth. The prediction is based on global mass balance, supplemented by a detailed geochemical and geophysical study of the region near the detector. The prediction is weakly dependent on mantle modeling. If BSE is correct, Uranium geo-neutrinos will produce between 25 and 35 events per year and 10^32 protons at Kamioka.

Geo-neutrino detection will determine the amount of long lived radioactive elements within our planet and fix the debated radiogenic contribution to the terrestrial heat. In addition, it will provide a direct test of the Bulk Silicate Earth model, a fundamental cosmochemical paradigm about the origin of the Earth. Unorthodox models of Earth's core (including the presence of potassium or the possibility of a giant reactor) can also be checked. This short review presents status and prospects of the field.

In preparation to the experimental results which will be available in the future, we consider geo-neutrino production in greater detail than in [F. Mantovani et al., arXiv:hep-ph/0309013], putting the basis for a more refined model. We study geo-neutrino production for different models of matter circulation and composition in the mantle. By using global mass balance for the Bulk Silicate Earth, the predicted flux contribution from distant sources in the crust and in the mantle is fixed within +-15% (full range). A detailed geological and geochemical investigation of the region near the detector has to be performed, for reducing the flux uncertainty from fluctuations of the local abundances to the level of the global geochemical error. A five-kton detector operating over four years at a site relatively far from nuclear power plants can measure the geo-neutrino signal with 5% accuracy (1 sigma). It will provide a crucial test of the Bulk Silicate Earth and a direct estimate of the radiogenic contribution to terrestrial heat.