Evapotranspiration over agroforestry sites in Germany

Abstract. In past years the interest in growing crops and trees for bioenergy production increased. One agricultural practice is the mixed cultivation of fast growing trees and annual crops or perennial grass-lands on the same piece of land, referred to as one type of agroforestry. The inclusion of tree strips into the agricultural landscape has been shown – on the one hand – to lead to reduced wind speeds and higher carbon sequestration above-ground and in the soil. On the other hand, concerns have been risen about increased water losses to the atmosphere via evapotranspiration (ET). Therefore it was our main objective to proof if agroforestry systems have higher ET compared to monoculture systems. We followed a replicated measurement design to investigate the impact of agroforestry (AF) on ET under consideration of different ambient conditions. We measured actual ET at five agroforestry sites in direct comparison to five monoculture (MC) sites in Northern Germany in 2016 and 2017. We used an eddy covariance energy balance set-up (ECEB) and a low-cost eddy covariance set-up (EC-LC) to measure actual evapotranspiration over each agroforestry and each monoculture system. We conducted direct eddy covariance (EC) measurement campaigns of approximately four weeks duration for method validation. Results from the short-term measurement campaigns showed a high agreement between ET from EC-LC and EC, indicated by slopes of a linear regression analysis between 0.86 and 1.3 (R2 between 0.7 and 0.94) across sites. Root mean square errors of LE from EC-LC vs. EC were half as small as from ECEB vs. EC, indicating a superior performance of EC-LC compared to ECEB. The overall effect of agroforestry on system-scale ET for the two years was small compared to the monoculture systems. Differences between annual ET over AF and MC from the two years and both measuring set-ups were not significant (p = 0.3557). We interpret this as an effect of compensating small-scale differences in ET when ET is measured on system-scale. A reduction of ET is expected to be strongest next to the tree strips due to reduced wind speed and limited incident radiation relative to an open field. Whereas in between the tree strips ET is expected to increase due to higher incident radiation. Most likely differences in ET rates next to and in between the tree strips are of the same order of magnitude and compensate each other on system scale. In contrast, we found a strong dependency of ET on the local climate, characterized by the evapotranspiration index (∑ET/precipitation). We observed significant (p = 0.0007098) higher mean evapotranspiration indices across sites for a drier than normal year (2016) compared to a wet year (2017) independent of the land-use or method. We conclude that agroforestry has not resulted in an increased water loss to the atmosphere indicating that agroforestry in Germany can be a land-use alternative to conventional agriculture.