Advanced monitoring of agroecosystem at risk of salinization as an adaptation measure to climate change

Adaptation measures to climate change demand developing an advanced system for monitoring, predicting and reporting on the state on various ecosystems. This especially refers to river valleys, estuaries and deltas that are predominantly used for agriculture and at the same time are at high risk of various degradations due to climate change. In order to control risks, it is important to obtain spatially and temporally articulated data on soil quality. Although the problem of seawater intrusion and salinization of surface and groundwater occurs in all of the Croatian coastal areas, hence increasing the risk of soil degradation, these processes are most evident in the area of Neretva river delta which is considered one of the most vulnerable areas to climate change in Croatia. To assess soil salinity risk in this area, seasonal soil monitoring was set (from 2009-now) with sampling campaigns at the end of the dry/wet season. In 2020, the soil monitoring network was upgraded with frequency domain sensors (FDR) to continuously record changes in soil. The aim of this research was to asses soil salinity dynamics taking into account different monitoring approaches and different temporal resolution of data collecting. Two soil monitoring stations (SMS) have been set up at two different locations with a pronounced risk of salinization in Neretva river delta. Two SMS differ regarding depth of gleyic horizon, soil texture, organic matter content, salinity level and land use: SMS1 vegetable production, SMS2 citrus orchard. At each monitoring station extensive characteristisation of soil physical and chemical characteristics was conducted. Frequency domain sensors measuring bulk electrical conductivity (ECb) were installed at 4 different depths (0-25 cm ; 25-50 cm ; 50-75 cm ; 75-100 cm) with 10 minutes measuring interval. Within the research area an automated weather station was installed (Pinova Meteo weatherstation) to record precipitation and other relevant meteo data. According to the long term seasonal soil monitoring in the wet season the soil was non saline at SMS1 with average value of 1, 06 dS/m and slightly saline at SMS2 with average of 3, 17 dS/m. Maximum values recorded with this type of monitoring were 4, 79 dS/m at SMS1 and 4, 95 dS/m at SMS2 respectively. FDR salinity monitoring was set at the beginning of November 2020 and in this paper data for the first 70 days of monitoring were analysed. Within this time frame at SMS1 (1 m depth) average ECb was 2, 86 dS/m with an absolute maximum value of 6, 38 dS/m (50-75 cm). Average ECb at SMS2 (1 m depth) was 4, 07 dS/m with absolute maximum of 6, 21 dS/m (75-100 cm). Higher coefficient of variation of ECb was recorded at SMS1 in all horizons in regard to SMS2 where ECb variation was very low, raging from 3, 7 % (75-100 cm) to 14, 06 % (0-25). Highest coefficients of variability at SMS1 were recorded in horizon 75- 100 cm (44, 79 %) and 0-25 cm horizon (41, 32 %) indicating high soil salinity variation. However SMS2 which had higher ECb in each horizon displayed temporal stability with average coefficient of variation of ECb up to 1 m of 10 %. At SMS1 average coefficient of variation was over 33 % showing temporal instability of ECb. It is important to emphasize that this is an initial set of data and firm conclusions about the dynamics of salinization cannot yet be made but they suggest the importance of developing site specific soil salinity mitigation actions.