BIBLIOTECA DE SIGNOS

7.2.1. Material and Methods

Along the valuable habitats of the Biebrza Valley, the ones of special importance are fens (both well conserved and degraded) for which the groundwater recharge (and thus – discharge) remains a crucial process that sustain water availability. Therefore, one of the principal goals in climate-change impact assessment on the area was to assess the prospective changes in groundwater recharge within the Biebrza Valley. In order to do so, we used a set of high-resolution regional climate simulations performed by several state-of-art RCMs (regional climate models) and driven by different GCMs (global climate models) in the framework of the EU-FP6 ENSEMBLES project and consider 8 GCM/RCM combinations all for the SRES A1B emission scenario. The spatial resolution of the RCM data is ~ 25km. Those data sets have been selected out of the ENSEMBLE matrix, with the criterion that they come with all required parameters for further hydrological analysis and are available from 1961 until 2100. No bias correction has been performed. Instead, when doing the investigation of scenario impacts, scenario data of the years 1961-2000 serve as the reference conditions and are compared against scenario data of the years 2061-2100 (scenario period).

Figure 41. The hydrological processes of the SWIM model including the parameter demand.

The eco-hydrological model SWIM The model SWIM has been chosen for this study because it integrates the relevant hydrological and plant processes necessary to investigate the impacts of climate change on water resources like evapotranspiration, percolation, surface runoff, interflow, groundwater recharge, plant water uptake, vegetation dynamics and river routing (Krysanova et al. 1998, Hattermann et al. 2005). Important is the combination of hydrological and plant processes in one model, because changes in climate also directly affect plant growth and phenology and feedback on the water cycle e.g. by water uptake over the vegetation season. A three-level scheme of spatial disaggregation from basin to subbasins and to hydrotopes is used. Water fluxes, plant growth and nutrient dynamics are calculated for every element on a daily time step, where up to 10 vertical soil layers can be considered. The outputs from the elements are aggregated at the subbasin scale, taking water and nutrient retention into account. The lateral fluxes are routed over the river network, considering transmission losses. However, in this study only the vertical processes where considered as groundwater recharge was the target component of the water balance. The main hydrological processes in SWIM are shown in Figure 1. A comprehensive description of the model can be found in Krysanova et al. (1998). The Turc-Ivanov method is used to estimate the potential evapotranspiration. Soil evaporation and plant transpiration are calculated as functions of leaf area index LAI using the approach of Ritchie (1972). Surface runoff is determined using a modification of the Soil Conservation Service (SCS) curve number technique. Water, which has infiltrated into the soil, percolates through the soil layers using a storage routing technique. The water percolated from the bottom soil layer, which reaches the groundwater table with some delay time, is defined as groundwater recharge (Fig. 40).

[56]

The Turc-Ivanov method is used to estimate the potential evapotranspiration. Soil evaporation and plant transpiration are calculated as functions of leaf area index LAI using the approach of Ritchie (1972). Surface runoff is determined using a modification of the Soil Conservation Service (SCS) curve number technique. Water, which has infiltrated into the soil, percolates through the soil layers using a storage routing technique. The water percolated from the bottom soil layer, which reaches the groundwater table with some delay time, is defined as groundwater recharge (Fig. 1).

A simplified EPIC approach (Williams et al., 1984) is included in SWIM for simulating arable crops (like wheat, barley, rye, maize, potatoes) and aggregated vegetation types (e.g. 'mixed forest') on a daily time step, using specific parameter values for each crop/vegetation type. The potential increase in biomass is adjusted each day if one of the plant stress factors is less than 1, considering stresses caused by water and temperature. The water stress factor is calculated by comparing water supply in soil and water demand, assuming that about 30 % of the total water comes from the top 10 % of the root zone. The approach allows roots to compensate for water deficits in certain layers by using more water in other layers with adequate supply. Different model extensions have been implemented to account for specific hydrological features of the different basins. This concerns firstly wetland processes having a high impact on hydrological processes in lowland basins with shallow groundwater (cf. Hattermann et al. 2008). Additional uptake of water is possible to a certain extend by vegetation in riparian zones and wetlands, where the plant roots reach the groundwater table. The model SWIM was set-up for two typical soils located in the area, a histosol and a podzol, the first one a loamy soil with a high organic content and connected to groundwater, the second one a sandy soil with high water conductivity without groundwater connection. The vegetation cover considered in the simulations is grassland. SWIM was driven for the entire period 1961-2100 by daily climate data of precipitation, radiation and temperature (min., max., mean) of the respective regional models. Afterwards, changes in groundwater recharge have been calculated in terms of long-term daily differences 2061-2100 minus 1961-2000.

7.2.2. Results

Figures 42 and 43 give the vertical hydrological flows for two typical soils, an upland sandy Podzol without groundwater connection and a wetland Histosol connected to groundwater. Shown is the long-term daily average 1961-2000, in this example driven by climate data of the regional climate model REMO (with ECHAM5 GCM data as boundary condition).

The results illustrate that the dynamics of groundwater recharge are quite different with and without connection of plants to groundwater. Under wetland conditions, plants have access to additional water uptake from groundwater if soil water is exhausted (see Fig. 43). This effect is the highest in late summer. During this period, wetlands are areas of groundwater depletion and function as sinks in the water balance. Groundwater depletion in summer is relatively high because “actively” driven by plant uptake.

Figure 42. Vertical hydrological flows and LAI development for Podzol without groundwater connection in terms of the long-term daily average 1961-2000 (Prec – precipitation, Perc – groundwater recharge, EtPot – potential evapotranspiration, EtAct – actual

evapotranspiration, LAI – Leave Area Index).

Figure 43. Vertical hydrological flows and LAI development for Histosol with groundwater

connection in terms of the long-term daily average 1961-2000 (Prec – precipitation, Perc – groundwater recharge, EtPot – potential evapotranspiration, EtAct – actual

[58]

The results illustrate that the dynamics of groundwater recharge are quite different with and without connection of plants to groundwater. Under wetland conditions, plants have access to additional water uptake from groundwater if soil water is exhausted (see Fig. 43). This effect is the highest in late summer. During this period, wetlands are areas of groundwater depletion and function as sinks in the water balance. Groundwater depletion in summer is relatively high because “actively” driven by plant uptake. Groundwater recharge during winter is relatively low because of the low conductivity of the loamy soils and the high water holding capacity. When looking at the hydrological processes in the upland Podzol soil, groundwater recharge is zero during summer indicating that all water is used by plants. Additional uptake from groundwater is not possible. The period of groundwater replenishment is the late winter when soil water content reaches field capacity and a relatively high amount of water percolates to deeper soil layers because of the high conductivity and low water holding capacity of the soil.

7.2.3. Changes in groundwater recharge under scenario conditions

Figure 44 illustrates the impact of climate change on groundwater recharge in the target area. In all climate runs, total annual precipitation increases with partly decreases in summer and increases in winter. This leads to a total increase of groundwater recharge although plant demand may increase during summer because of the warmer conditions. In the upland site (sandy Podzol), groundwater recharge decreases especially in spring when plants start growing earlier and thus start taking up water earlier in the year under warmer conditions.

Figure 44. Changes in groundwater recharge for Podzol without groundwater connection as the long-term daily difference 2061-2100 minus 1961-2000 simulated by SWIM and driven by 8 different climate simulations. The possible increase in evapotranspiration under warmer climate conditions is in the target area overcompensated by the increase in annual precipitation leading to an increase in groundwater recharge. The increase is higher in upland areas and lower in wetlands and riparian zones because of the higher conductivity and low water holding capacity of the soils.

Figure 45. Changes in groundwater recharge for Histosol with groundwater connection as the long- term daily difference 2061-2100 minus 1961-2000 simulated by SWIM and driven by 8 different climate simulations. The possible increase in evapotranspiration under warmer climate conditions is in the target area overcompensated by the increase in annual precipitation leading to an increase in groundwater recharge. The increase is higher in upland areas and lower in wetlands and riparian zones because of the higher

conductivity and low water holding capacity of the soils.

In summer, changes are almost zero as recharge was very low anyway under reference conditions (see Figure 42) and additional uptake from groundwater is not possible. Under wetland conditions, comparable pattern can be seen with a distinct increase in groundwater recharge also in summer when additional uptake by plants from groundwater is decreased because of the in total higher water availability because of the total increase in precipitation and high water storage of the wetland soils.

7.3. Groundwater levels

The conclusions from the research on groundwater recharge given in the chapter 7.2 let to suspect that if external factors aimed at water level decline (drainage) do not occur, then the general increase of groundwater levels in the Biebrza Valley is likely to be expected in the future. This prognosis finds confirmation in field observations in recent years (Fig. 34, 35 and 38), which reflect the ongoing and long-term prospective trends in precipitation and temperature distribution. As such, this observation can become a starting point on discussion, whether the broad-scale habitat restoration projects aimed at increase of water levels are indispensible, or if the projected climatic variability is capable to speed-up the process of habitat re-wetting. Clearly the drainage should be limited to the smallest extent possible and it is desirable to reduce baseflow from degraded wetland areas by blocking ditches and re-establishment of abandoned (in result of technical measures) river channels, but it is likely that ongoing processes of increasing and stabilisation of groundwater recharge in the scale of Biebrza Valley is a good prerequisite to groundwater-dependent wetlands for their future development and evolution.

[60]

Similarly to the flood extent dynamics influenced by the prospective climate change it is also suspected, that groundwater levels – if not affected neither by the drainage nor the artificial water management in the upper-located parts of the Biebrza Catchment – will increase in the Biebrza Valley. This will result with any possibility in increasing saturation of soils. Within the natural and well preserved peatland (such as the Szuszalewo Fens), the water level increase will result in limitation of trees and shrubs development. Thus it will affect the contemporary strategies aimed at keeping the open landscapes of fens. Possibly, the long term projects aimed at trees and shrubs removal will require to be revisited and the prospective climatic conditions will have to be considered in the environmental management adaptation. Within the degraded groundwater dependent wetlands (the western part of the Middle Biebrza Basin) increasing groundwater levels in decomposed peat soils is likely to entail the nutrient migration from the sorption complex of the muck and moorsh soils to groundwater and adjacent rivers. One can therefore expect, that the eutrophication of degraded habitats and aquatic ecosystems will occur, which will entail the increased development of aquatic macrophytes. This process is likely to occur as a feedback with the summer flooding. However, without the direct data on soils and groundwater flow and discharge, such analyses have only a speculative dimension. However, this processes will have to be considered whilst the management adaptation of groundwater dependent ecosystems was started.