HS2.1.5

To meet the sustainable development goals and enable protection of surface waters, there is a strong need to plan and align forest management with the needs of the environment. The number one tool to succeed in sustainable spatial planning is accurate and detailed maps. High resolution soil moisture mapping over spatial large extent remains a consistent challenge despite its substantial value in practical forestry and land management. Here we present a novel technique combining LIDAR-derived terrain indices and machine learning to model soil moisture at 2 m spatial resolution across the Swedish forest landscape with high accuracy. We used field data from about 20,000 sites across Sweden to train and evaluate multiple machine learning (ML) models. The predictor features included a suite of terrain indices generated from national LIDAR digital elevation model and other ancillary environmental features, including surficial geology, climate, land use information, allowing for adjustment of soil moisture maps to regional/local conditions. In our analysis, extreme gradient boosting (XGBoost) outperformed the other tested ML methods (Kappa = 0.69, MCC= 0.68), namely Artificial Neural Network, Random Forest, Support Vector Machine, and Naïve Bayes classification. The depth to water index, topographic wetness index, and wetlands derived from Swedish property maps were the most important predictors for all models. With the presented technique, it was possible to generate a multiclass model with 3 classes with Kappa and MCC of 0.58. Besides the classified moisture maps, we also investigated the potential of producing a continuous map from dry to wet soils. We argue that the probability of a pixel being classified as wet from the 2-class model can be used as an index of soil moisture from 0% – dry to 100% – wet and that such maps hold more valuable information for practical forest management than classified maps.

The soil moisture map was developed to support the need for land use management optimization by incorporating landscape sensitivity and hydrological connectivity into a framework that promotes the protection of soil and water quality. The soil moisture map can be used to address fundamental considerations, such as;

• (i) locating areas where different land use practices can be conducted with minimal impacts on water quality;
• (ii) guiding the construction of vital infrastructure in high flood risk areas;
• (iii) designing riparian protection zones to optimize the protection of water quality and biodiversity.

How to cite: Ågren, A. M., Larson, J., Paul, S. S., Laudon, H., and Lidberg, W.: Combining LIDAR-derived Digital Terrain Indices and Machine Learning, for High Resolution National-scale Soil Moisture Mapping of the Swedish Forest Landscape., EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-909, https://doi.org/10.5194/egusphere-egu21-909, 2021.

Rainforests have been a major controller of local and global climate by maintaining carbon stocks and regulating global water cycle. However, the water cycle is increasingly impacted by climate change and ongoing deforestation, which forces rainforest ecosystems to adapt differently to increasing water-stress. To understand future rainforest dynamics towards changing hydroclimate, their resilience capacity to future changes and estimating potential tipping points, it is detrimental that we quantify moisture available to vegetation. However, due to the physical limitations in quantifying subsurface moisture availability of terrestrial ecosystems at continental scales, only rainfall is considered a primary control variable to represent the forest's ecohydrological status. In the present study, using remote-sensing derived rootzone storage capacity (S_r), we analyze the water-stress and drought coping strategies along rainforest-savanna transects in South America and Africa at different tree cover densities. We further classified the ecosystem's adaptability to water-stress into four classes: lowly, moderately, highly water-stressed forest, and savanna-grassland regime using empirical and statistical analysis. Based on these analyses, we can show that forests subsequently invest in their rooting strategy and modify their above-ground forest cover in response to the water-stress experienced by it. We observed that remote sensing-based rootzone storage capacity reveals important subsoil forest dynamics and can act as an important hydroclimatic stress indicator for vegetation. Monitoring of rootzone storage capacity helps open new paths to understanding the eco-hydrological state, ecosystem resilience, and adaptation dynamics in a rapidly changing climate.

Source: Singh, C. et al. (2020). Rootzone storage capacity reveals drought coping strategies along rainforest-savanna transitions. Environmental Research Letters, 15(12), 124021. doi: 10.1088/1748-9326/abc377

How to cite: Singh, C., Wang-Erlandsson, L., Fetzer, I., Rockström, J., and van der Ent, R.: Water stress and their implications on the ecohydrology of rainforests, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-12345, https://doi.org/10.5194/egusphere-egu21-12345, 2021.

In a recent paper we investigated how different catchment and climate properties influence transit time distributions. This was done by employing a physically-based spatially explicit 3D model in a virtual catchment running many different scenarios with different combinations of catchment and climate properties. We found that the velocity distribution of water fluxes through a catchment is more sensitive to certain properties while other factors appear less relevant. Now we expanded the approach by adding vegetation to the model and thus introducing new hydrologic processes (transpiration and evaporation) to the simulated water cycle. On the one hand we wanted to know how these new processes would influence transit times of the water fluxes to the stream, on the other hand we were interested in how exactly differences in the vegetation itself (e.g. rooting depth and leaf area index) would alter the various flux velocities (including transit times of transpiration and evaporation). It was very interesting to observe that streamflow in forested areas appeared to become older on average. We also found that transpiration was generally younger if the vegetation had shallower roots and/or a larger leaf area index. The biggest difference in the age of evaporation was detected for different amounts of subsequent precipitation (evaporation was generally younger in a wetter climate). In conclusion, we found that forests influence the age of the different water fluxes within a catchment. According to our results the overall hydrologic cycle is decelerated when adding vegetation to a model that otherwise only simulates evaporation.

Still, in order to make meaningful predictions on the age of hydrologic fluxes, it is not constructive to single out specific catchment and climate properties. The multitude of influences from different parameters makes it very challenging to find rules and underlying principles in the integrated catchment response. Therefore it is necessary to look at the individual parameters and their potential interactions and interdependencies in a bottom-up approach.

How to cite: Heidbüchel, I., Yang, J., and Fleckenstein, J. H.: How forests change transit times of transpiration, evaporation and streamflow – a modeling approach, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-11991, https://doi.org/10.5194/egusphere-egu21-11991, 2021.

Natural Flood Management (NFM) seeks to utilise natural processes within the landscape to reduce flood risk and is increasingly being viewed as a sustainable, cost effective, and complementary addition to flood defence infrastructure.  One NFM measure is to increase the proportion of forested lands within catchments draining to the communities at risk. Tree cover has good potential to reduce flood risk by increasing canopy evaporation, enhancing below and above ground flood storage and slowing the flow of water towards streams.  However, the extent to which these mechanisms are superior for forestry, compared to other land uses, and how they vary throughout the year and for different forest types remains difficult to predict, which is a major gap in our ability to quantify how forest cover can help reduce flood risk.

Here, we present a study that utilises LoRaWAN, a developing wireless sensor network technology, to provide real time collection of canopy interception and streamflow data at the Pennal catchment in Wales, UK.  LoRaWAN is an emerging Low Power Wide Area Network (LPWAN) protocol designed for Internet of Things (IoT) applications. The capability of LoRaWAN to operate under harsh attenuation and interference conditions make it well suited to the forest catchment area which is characterised by dense vegetation and varied topography.

This study will utilise a network of tipping bucket rain gauges and stream flow monitors distributed in different forest types and densities. The rain gauges and water level monitors are the end devices (IoT things) in the network which perform a direct communication with LoRaWAN Gateways, from which the data is ‘pushed’ to a server for storing and assimilation. The data will be used to develop and validate a coupled canopy and soil hydrology model.  This will guide forest management and aid in quantifying the effects of natural flood management techniques, initially within the Pennal catchment, with a view to expanding to the regional scale.

How to cite: Cooper, M., Dickens, M., Patil, S., and Thomas, H.: Developing a novel monitoring system to determine rainfall interception from different forest types, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-15950, https://doi.org/10.5194/egusphere-egu21-15950, 2021.

The hysteretic behavior between soil moisture and streamflow has received only little attention in the context of hillslope hydrological processes, despite the overarching role it plays in the understanding of the temporal and spatial dynamics of hillslope responses. In this study, hydro-meteorological data were collected daily on bi-hourly basis from 2009 to 2013 over 56 soil moisture measuring points at various depths (10, 30, and 60 cm) with 147 distinct storm events chosen for investigation. A bivariate analysis approach was implemented to characterize 8,232 hysteretic behaviors between streamflow and soil moisture with a view to exploring its patterns and uniformities using data obtained in the following timescale - the whole period of campaign, seasonally and storm event. In addition, hydrological control features such as antecedent soil moisture, rainfall intensity and duration, soil depth and hillslope positions were examined to establish the degree of control it poses on hillslope responses. Our investigation showed three dominant responses – clockwise, counter-clockwise and no response. Clockwise response which implied that streamflow peaked before soil moisture, governed the entire period of campaign with the frequency of responses significantly decreasing as depth increases, except for some downslope points located around the riparian zone. Furthermore, distinct variation in the hysteretic behavior of the hillslope under seasonal timescale was observed, with clockwise responses dominating summer and fall season whereas counter clockwise responses prevailed in the spring season. Our findings further reveals that antecedent soil moisture condition and soil depth were the major drivers that influenced the general response of the hillslope.

How to cite: Bassey Friday, B., Lee, E., and Kim, S.: Soil Moisture and Streamflow Relationship in Forested Hillslope: A Perspective on their Hysteretic Behavior., EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-10525, https://doi.org/10.5194/egusphere-egu21-10525, 2021.

Forest fires are becoming more frequent and larger in the subarctic Canadian Shield, so understanding the effect of fire on catchment scale water budgets is becoming increasingly important.  The objective of this study was to determine the water budget impact of a forest fire that partially burned a ~450 km² subarctic Canadian Shield basin.  Water budget components were measured in a pair of catchments; one burnt and another unburnt. Burnt and unburnt areas had comparable net radiation, but ground thaw was deeper in burned areas.  Snowpacks were deeper in burns. Differences in streamflow between the catchments were within measurement uncertainty.  Enhanced winter streamflow from the burned watershed was evident by icing growth at the streamflow gauge location, which was not observed in the unburned catchment.  A new framework to assess hydrological resilience to forest fire across the region revealed that watersheds with higher bedrock and open water fractions are more resilient to hydrological change after fire in the subarctic shield, and resilience decreases with increasingly wet conditions.  

How to cite: Spence, C., Hedstrom, N., Tank, S., Quinton, W., Olefeldt, D., Goodman, S., and Dion, N.: Hydrological resilience to forest fire in the subarctic Canadian Shield, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-6419, https://doi.org/10.5194/egusphere-egu21-6419, 2021.

Countless studies have demonstrated ways in which forests and trees affect catchment water balances. Water balance differences between forested and non-forested landscapes are often attributed to characteristics related to trees’ ability to take up and transpire water, as well as their ability to intercept precipitation. However, another potentially important characteristic of forests that has been largely overlooked in hydrologic studies is the retention and accumulation of debris, litter and deadwood on the forest floor. Here we leverage ongoing measurements at the new hillslope laboratory “Waldlabor” in Zurich, Switzerland, where water retention in forest litter, deadwood and the top soil layer has been investigated using frequent field campaigns and innovative new sensing techniques.

Several approaches were used to determine the maximum storage capacity as well as the storage dynamics of different types and layers of litter. In-lab saturation experiments revealed the maximum storage capacity of various litter types (i.e., leaf and needle litter). Those values were also supported with field pre- and post- rainfall sampling campaigns to determine in-situ litter storage dynamics, as well as to understand the interplay between litter interception and soil-water recharge. Importantly, recharge was often substantially smaller at plots with litter, compared to those without litter. The storage and water retention capacity of deadwood samples was measured in the field by logging the diurnal differences in deadwood weight over a six month period. Dew and fog deposition during the night led to larger water availability for evaporation during the day. We measured increased humidity at sensors in the forest at 1 and 3m heights respectively, compared to the humidity outside the forest. Daily weight measurements over eight weeks of 40 deadwood pieces at our forest site revealed differences in the storage capacity depended on the degree of decomposition. Additionally, we found that water stored in forest floor spruce cones (daily measurements of 20 pieces) actively contributed to evaporation fluxes.

The combination of continuous sensor measurements (soil moisture, deadwood water content), field measurements (litter and deadwood grab samples) as well as laboratory work (saturation experiments) revealed the water storage and retention capacity of litter and deadwood in a typical temperate mixed forest and their contribution to evaporation. These measurements are one component of the new ETH Zürich “Waldlabor” research infrastructure, which also includes measurements of precipitation, xylem water, soil water, groundwater, and discharge amounts, isotope ratios, and other chemical characteristics.

How to cite: Floriancic, M. G., Allen, S. T., and Molnar, P.: Litter and deadwood water retention processes in a temperate mixed forest, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-15217, https://doi.org/10.5194/egusphere-egu21-15217, 2021.