A reliable estimation of primary production of terres-tions in plant water stress, temperature, leaf aging, and processes such as respiration and changes in allocation trial ecosystems is often a prerequisite for land survey and management, while being important also in ecologi-pattern between above-and below-ground compartments.
Over the 3 study years, averaged values of e g , e n , and e an cal and climatological studies. At a regional scale, grassland primary production estimates are increasingly being were found to be 1.92, 0.74, and 0.29 g DM (MJ IPAR) Ϫ1 , respectively. e g and e n exhibited large interannual and made with the use of satellite data. In a currently used approach, regional gross, net, and aboveground net pri-seasonal variations mainly due to changes in water limitations during the growing season. Interannual variations mary productivity (GPP, NPP, and ANPP) are derived from the parametric model of Monteith and are calcu-of e an were much less important. However, for shorter periods, e an exhibited very contrasting values from regrowth lated as the product of the fraction of incident photosynthetically active radiation absorbed by the canopy (f APAR ) to senescence. The calculation of ANPP seems less prone to errors due to environmental effects when computed on and gross, net, and aboveground net production (radiation-use) efficiencies (e g , e n , and e an ); f APAR being derived an annual basis. When estimating GPP and NPP, better results are expected if water limitations are taken into ac-from indices calculated from satellite-measured reflectances in the red and near infrared. The accuracy and count. This could be possible through the estimation of a water-stress factor by using surface temperature or other realism of the primary production values estimated by this approach therefore largely depend on an accurate es-indices derived from thermal infrared remote sensing data. The limitations due to temporally varying efficienc-timation of e g , e n , and e an . However, data are scarce for production efficiencies of semiarid grasslands, and their ies, shown here for shortgrass ecosystems, are also relevant to all drought-exposed ecosystems, particularly those time and spatial variations are poorly documented, often with abundant evergreen or perennial species. Publeading to large errors for the estimates. In this paper, a lished by Elsevier Science Inc. modeling approach taking into account relevant ecosystem processes and based on extensive field data was used to estimate time variations of e g , e n and e an of a shortgrass site in Arizona. These variations were explained by varia-