Evaluating soil sodium indices in soils of volcanic nature conducive or suppressive to Fusarium wilt of banana

The ability of three soil Na indices to predict soil conduciveness or suppressiveness to disease caused by the soil fungus Fusarium oxysporum f. sp. cubense was evaluated in seven banana plantations from the Canary Islands (Spain). These indices were exchangeable sodium percentage (ESP), soluble Na (SS0) and sodium adsorption ratio (SAR0) in 1:2.5 soil–water extracts (SARw and total cationic concentration (TCCw)=0. Sodium selectivity coefficients (KG0,K0) and TCC0 were calculated from soil exchange and solution data. The effects of ESP, SAR0, SS0, TCC0, KG0 and K0 on soil available iron (Fe extracted from soil by DTPA) and aggregate stability in water (water-stable aggregates (WSA), 200–2000 μm) were also studied. Our results showed that SAR0 calculated using cationic concentrations in 1:2.5 extracts might be a good indication of a relationship between SS0 and soluble divalent cations in conducive and suppressive volcanic soils to Fusarium. Both TCC0 and dispersion–flocculation concentrations seem to be not linked to soil suppressiveness or conduciveness to Fusarium wilt. These results suggested that soil physical properties seem to be not controlled by Na behaviour in these type of soils and, therefore, sodicity and salinity should not be a problem from a physical point of view. Moreover, SS0 and SAR0 were always greater in suppressive areas than in conducive areas. SAR0 was significantly correlated with SS0 but correlations between ESP against SS0 and SAR0 were weak. For SAR0 values above 2.5 (mmolc l−1)1/2 and ESP values below 15%, the exchangeable Na did not seem to be related to the capacity of suppressive areas to release more Na to soil solution. Larger values of SS0 were observed in suppressive areas for these values of SAR0 and ESP. It implies a lower quantity of soluble Na salts in conducive samples. A high Na salt content in soil can produce an increase of soil pH, which exerts a negative influence on available Fe release to soil solution. A clear separation between conducive and suppressive samples from relations between SS0 and SAR0 against WSA and Fe–DTPA showed that SS0 and SAR0 can be satisfactory indices to study the influence of Na concentrations on the incidence of Fusarium wilt. The mass of WSA increase in conducive areas might be favoured by the smaller amounts of soil solution Na found in these samples. In conclusion, our data provide evidence that release of Na to soil solution could favour soil suppressiveness to Fusarium wilt limiting soil aggregation and the availability of Fe, at least in soils of volcanic nature that are not affected by salinity or sodicity processes.