Respect-A2 From Field Scale Eco-Hydrological Process Understanding To Landscape Scale Water Fluxes

From A Hydrological Perspective, Many Land Surface Models (Lsm) Have Great Deficiencies In The Representation Of Water-Related Feedback Mechanisms Induced Through Biotic, Climatic, Topographic And Hydrologic Landscape Features. This Results In Oversimplified Description Of Hydrological Processes, Effecting The Credibility Of These Models In Simulating Water Fluxes (Scheiter, Langan & Higgins 2013; Clark Et Al. 2015). These Effects Are Especially Pronounced In Complex And Diverse Landscapes Like The Tropical Mountain Rainforest Of South Ecuador (Sakschewski Et Al. 2016). Building On Our Findings In The Phase I And The Overall Objectives Of The Research Group In Phase Ii, We Have Identified Three Objectives For This Proposal: (1) Improve Our Eco-Hydrological Knowledge Of Diverse Mountain Forest Systems, (2) Implement This Understanding Into Humbol-Td And Develop Methods For Upscaling Of Field Scale Eco-Hydrological Process To The Landscape Scale, And (3) Utilize The Revised Humbol-Td To Test The Central Hypotheses Of Respect Considering Land Use And Climate Change, As Well As The Feedbacks Between Soil Water Availability And Functional Trait Diversity. With Regard To The First Objective, We Will Extend Our Studies Of The Below Ground Heterogeneity Of Eco-Hydrological Processes In Mrf To Mdf. We Will Also Intensify Our Work On The Field Scale In Mrf Around The Different Flow Processes And Their Effects On The Composition Of The Stable Isotopes Of Water. For This Purpose, We Will Sample Soil Water At Different Extraction Pressures And Thus Quantify The Mobility Of The Soil Water. The Derived Improved Process Understanding Will Be Implemented And Tested In Lsmhydro (Djabelkhir Et Al. 2016) To Match Our Second Objective. This Better Eco-Hydrological Understanding Of Plant Water Uptake, Mixing Processes In The Vadose Zone And Runoff Generation Will Support Our Ambition To Scale Up From Fields To The Landscape. This Upscaling Requires A Sophisticated Balancing Between Trade-Offs Induced By Long Model Run Times On Large Spatial Domains, The Wish To Be As Realistic As Possible With Regard To Considered Hydrological Process Understanding And Achieving A Satisfactory Model Performance. As Part Of The Second Objective, We Will Therefore Set Up And Develop A Semi-Automatic Testing Of A Number Of Different Hydrological Model Structures In The Sense Of A Rejectionist Framework, Where We Finally Keep Those Structures On The Sub-Catchment Scale That Outperforms All Others. The Third Objective Will Include Finally The Regional Application Of Humbol-Dt, Its Application To Various Climate Change Scenarios As Well As Natural And Anthropogenic Replacement Systems, And The Detailed Analyses Of The Simulated Functional Trait Diversity By Lsmbio In Relation To The Simulated Soil Water Availability By Lsmhydro.