Effects of magnesium contents on the Freundlich adsorption isotherm constants and phosphorus availability in agricultural soils of Wallonia (Belgium) Cobert Florian 1 , Barbieux Sophie 1 , Renneson Malorie 1 , Colinet Gilles 1 1 Université de Liège, Gembloux Agro-Bio Tech. Unité Systèmes Sol-Eau, 2 passage des déportés. B- 5030 Gembloux Belgique. (florian.cobert@ulg.ac.be ; gilles.colinet@ulg.ac.be) The reserves of total phosphorus (P tot ) in Walloon soils are high, 904 mgP/kg on average in the topsoil [1]. It resulted from continuous and excessive fertilization and manure applications [2]. However, only 9% of P tot on average (71.6 mgP/kg) is available in Walloon topsoil due to high soil P sorption capacity [1]. Thus, a better understanding of P behavior in Walloon soils is needed to adapt practices of fertilization to the requirement of limiting and improving phosphorus (P) availability in soils of Wallonia. Recently, different studies have shown that during the last decade, magnesium (Mg) content in soils of Wallonia increased between 10 to 35% [3]. Hence, Mg contents could influence the availability of P in these same soils [4]. In that context we studied P available (PO 4 -P) in different agricultural soils of Wallonia (N=5) by batch equilibrium experiments in order to define: (1) the influence of Mg content on the Freundlich adsorption coefficients ; (2) the effects of Mg content on the magnitude of P adsorption. A 3 g air dried sample was weighed into a 100 mL centrifuge tube with a cap and equilibrated with 60 mL of one of six P solutions having 0, 0.1, 1, 10, 75, 150 ppm P. The solution added to the soil was prepared by dissolving KH 2 PO 4. Few drops NaNOH 3 0.01M was added. Mg solutions prepared by dissolving MgSO 4 were added to different centrifuge tubes to produce a range of Mg contents between 0 and 40%. To inhibit the microbial activity in each tube, three drops of chloroform were added. The tubes were shaken for 72h for equilibration purpose and subsequently were centrifuged and supernatant was collected for P determination. The P concentration in the extracts was determined by using the method of Murphy and Riley [5] with absorption measure using a spectrophotometer at a wavelength at 880 nm. The Mg content in the extracts was determined by using an ICP-AES. In a first step, the Freundlich adsorption isotherms were defined in five Walloon soils by batch equilibrium experiment. These results show that the addition of Mg influences the Freundlich adsorption coefficients K and 1/n in each soil. Thus, the application of Mg significantly increased the Freundlich K values and decreased 1/n compared to the control treatment in each soil. This suggests that more P was sorbed on soil and less was present in the soil solution. In a second step, the effects of Mg content on the magnitude of P adsorption in Walloon soils were determined. Our results show that the addition of Mg influences the PO 4 -P in soil solutions. Thus, more the addition of Mg increased, more the PO 4 -P in soil solution decreased. This suggests that addition of Mg decreased the PO 4 -P in Walloon soils tested. This study has shown that the Mg is an important parameter which governs P availability in agricultural soils of Wallonia. However, further studies are needed to understand the impact of Mg fertilization such as, Mg liming and struvite, on soil P availability. [1] Renneson, M. et al., 2013. Relationships between the P status of surface and deep horizons of agricultural soils under various cropping systems and for different soil types: a case study in Belgium. Soil Use and Management. 29: 103-113 [2] Ranatunga, T.D. et al., 2013. Phosphorus distribution in soil aggregate size fractions in a poultry litter applied soil and potential environmental impacts. Geoderma. 192: 446-452 [3] Genot, V. et al., 2012. Base de données sols de REQUASUD [4] Cobert, F. et al., 2013. Phosphorus availability in agricultural soils of Wallonia (Belgium) – A modeling approach. Mineralogical Magazine. Goldschmidt2013 Conference abstracts. 899