More Like this

1. Soil temperature and moisture as key controls of phosphorus export in mountain watersheds

   https://doi.org/10.1016/j.scitotenv.2024.170958  

   Gianniny, Gordon; Stark, John M; Abbott, Benjamin W; Lee, Raymond; Brahney, Janice (April 2024, Science of The Total Environment)

   Oligotrophic mountain lakes act as sensitive indicators of landscape-scale changes in mountain regions due to their low nutrient concentration and remote, relatively undisturbed watersheds. Recent research shows that phosphorus (P) concentrations are increasing in mountain lakes around the world, creating more mesotrophic states and altering lake ecosystem structure and function. The relative importance of atmospheric deposition and climate-driven changes to local biogeochemistry in driving these shifts is not well established. In this study, we test whether increasing temperatures in watershed soils may be contributing to the observed increases in mountain lake P loading. Specifically, we test whether higher soil temperatures increase P mobilization from mountain soils by accelerating the rate of geochemical weathering and soil organic matter decomposition. We used paired soil incubation (lab) and soil transplant (field) experiments with mountain soils from around the western United States to test the effects of warming on rain-leachable P concentration, soil P mobilization, and soil respiration. Our results show that while higher temperature can increase soil P mobilization, low soil moisture can limit the effects of warming in some situations. Soils with lower bulk densities, higher pH, lower aluminum oxide contents, and lower ratios of carbon to nitrogen had much higher rain-leachable P concentration across all sites and experimental treatments. Together, these results suggest that mountain watersheds with high-P soils and relatively high soil moisture could have the largest increases in P mobilization with warming. Consequently, lakes and streams in such watersheds could become especially susceptible to soil-driven eutrophication as temperatures rise. 

   more » « less
2. Manipulative experiments demonstrate how long-term soil moisture changes alter controls of plant water use

   https://doi.org/10.1016/j.envexpbot.2017.12.010  

   Grossiord, Charlotte; Sevanto, Sanna; Limousin, Jean-Marc; Meir, Patrick; Mencuccini, Maurizio; Pangle, Robert E.; Pockman, William T.; Salmon, Yann; Zweifel, Roman; McDowell, Nate G. (August 2018, Environmental and Experimental Botany)
3. Soil moisture and vapor pressure deficit controls of longleaf pine physiology: results from a throughfall reduction study

   https://doi.org/10.1007/s00468-023-02423-3  

   Mendonca, Caren C.; Samuelson, Lisa J.; Stokes, Tom A.; Ramirez, Michael R.; Gonzalez-Benecke, Carlos; Aspinwall, Michael J. (August 2023, Trees)
4. A Reduced-Adjoint Variational Data Assimilation for Estimating Soil Moisture Profile from Surface Soil Moisture Observations

   https://doi.org/10.1109/IGARSS47720.2021.9554864  

   Heidary, Parisa; Farhadi, Leila; Altaf, Muhammad Umer (July 2021, International Geoscience and Remote Sensing Symposium)
5. Assessment of a Spatiotemporal Deep Learning Approach for Soil Moisture Prediction and Filling the Gaps in Between Soil Moisture Observations

   https://doi.org/10.3389/frai.2021.636234  

   ElSaadani, Mohamed; Habib, Emad; Abdelhameed, Ahmed M.; Bayoumi, Magdy (March 2021, Frontiers in Artificial Intelligence)

   Soil moisture (SM) plays a significant role in determining the probability of flooding in a given area. Currently, SM is most commonly modeled using physically-based numerical hydrologic models. Modeling the natural processes that take place in the soil is difficult and requires assumptions. Besides, hydrologic model runtime is highly impacted by the extent and resolution of the study domain. In this study, we propose a data-driven modeling approach using Deep Learning (DL) models. There are different types of DL algorithms that serve different purposes. For example, the Convolutional Neural Network (CNN) algorithm is well suited for capturing and learning spatial patterns, while the Long Short-Term Memory (LSTM) algorithm is designed to utilize time-series information and to learn from past observations. A DL algorithm that combines the capabilities of CNN and LSTM called ConvLSTM was recently developed. In this study, we investigate the applicability of the ConvLSTM algorithm in predicting SM in a study area located in south Louisiana in the United States. This study reveals that ConvLSTM significantly outperformed CNN in predicting SM. We tested the performance of ConvLSTM based models by using a combination of different sets of predictors and different LSTM sequence lengths. The study results show that ConvLSTM models can predict SM with a mean areal Root Mean Squared Error (RMSE) of 2.5% and mean areal correlation coefficients of 0.9 for our study area. ConvLSTM models can also provide predictions between discrete SM observations, making them potentially useful for applications such as filling observational gaps between satellite overpasses. 

   more » « less