Impact of calibrating a low-cost capacitance-based soil moisture sensor on AquaCrop model performance.

Sensor data and agro-hydrological modeling have been combined to improve irrigation management. Crop water models simulating crop growth and production in response to the soil-water environment need to be parsimonious in terms of structure, inputs and parameters to be applied in data scarce regions. Irrigation management using soil moisture sensors requires them to be site-calibrated, low-cost, and maintainable. Therefore, there is a need for parsimonious crop modeling combined with low-cost soil moisture sensing without losing predictive capability. This study calibrated the low-cost capacitance-based Spectrum Inc. SM100 soil moisture sensor using multiple least squares and machine learning models, with both laboratory and field data. The best calibration technique, field-based piece-wise linear regression (calibration r2 = 0.76, RMSE = 3.13 %, validation r2 = 0.67, RMSE = 4.57 %), was used to study the effect of sensor calibration on the performance of the FAO AquaCrop Open Source (AquaCrop-OS) model by calibrating its soil hydraulic parameters. This approach was tested during the wheat cropping season in 2018, in Kanpur (India), in the Indo-Gangetic plains, resulting in some best practices regarding sensor calibration being recommended. The soil moisture sensor was calibrated best in field conditions against a secondary standard sensor (UGT GmbH. SMT100) taken as a reference (r2 = 0.67, RMSE = 4.57 %), followed by laboratory calibration against gravimetric soil moisture using the dry-down (r2 = 0.66, RMSE = 5.26 %) and wet-up curves respectively (r2 = 0.62, RMSE = 6.29 %). Moreover, model overfitting with machine learning algorithms led to poor field validation performance. The soil moisture simulation of AquaCrop-OS improved significantly by incorporating raw reference sensor and calibrated low-cost sensor data. There were non-significant impacts on biomass simulation, but water productivity improved significantly. Notably, using raw low-cost sensor data to calibrate AquaCrop led to poorer performances than using the literature. Hence using literature values could save sensor costs without compromising model performance if sensor calibration was not possible. The results suggest the essentiality of calibrating low-cost soil moisture sensors for crop modeling calibration to improve crop water productivity.

Integration of Remote Sensing and Mexican Water Quality Monitoring System Using an Extreme Learning Machine.

Remote Sensing, as a driver for water management decisions, needs further integration with monitoring water quality programs, especially in developing countries. Moreover, usage of remote sensing approaches has not been broadly applied in monitoring routines. Therefore, it is necessary to assess the efficacy of available sensors to complement the often limited field measurements from such programs and build models that support monitoring tasks. Here, we integrate field measurements (2013-2019) from the Mexican national water quality monitoring system (RNMCA) with data from Landsat-8 OLI, Sentinel-3 OLCI, and Sentinel-2 MSI to train an extreme learning machine (ELM), a support vector regression (SVR) and a linear regression (LR) for estimating Chlorophyll-a (Chl-a), Turbidity, Total Suspended Matter (TSM) and Secchi Disk Depth (SDD). Additionally, OLCI Level-2 Products for Chl-a and TSM are compared against the RNMCA data. We observed that OLCI Level-2 Products are poorly correlated with the RNMCA data and it is not feasible to rely only on them to support monitoring operations. However, OLCI atmospherically corrected data is useful to develop accurate models using an ELM, particularly for Turbidity (R2 = 0.7). We conclude that remote sensing is useful to support monitoring systems tasks, and its progressive integration will improve the quality of water quality monitoring programs.

Laboratory Calibration and Performance Evaluation of Low-Cost Capacitive and Very Low-Cost Resistive Soil Moisture Sensors.

Soil volumetric water content ( V W C ) is a vital parameter to understand several ecohydrological and environmental processes. Its cost-effective measurement can potentially drive various technological tools to promote data-driven sustainable agriculture through supplemental irrigation solutions, the lack of which has contributed to severe agricultural distress, particularly for smallholder farmers. The cost of commercially available V W C sensors varies over four orders of magnitude. A laboratory study characterizing and testing sensors from this wide range of cost categories, which is a prerequisite to explore their applicability for irrigation management, has not been conducted. Within this context, two low-cost capacitive sensors-SMEC300 and SM100-manufactured by Spectrum Technologies Inc. (Aurora, IL, USA), and two very low-cost resistive sensors-the Soil Hygrometer Detection Module Soil Moisture Sensor (YL100) by Electronicfans and the Generic Soil Moisture Sensor Module (YL69) by KitsGuru-were tested for performance in laboratory conditions. Each sensor was calibrated in different repacked soils, and tested to evaluate accuracy, precision and sensitivity to variations in temperature and salinity. The capacitive sensors were additionally tested for their performance in liquids of known dielectric constants, and a comparative analysis of the calibration equations developed in-house and provided by the manufacturer was carried out. The value for money of the sensors is reflected in their precision performance, i.e., the precision performance largely follows sensor costs. The other aspects of sensor performance do not necessarily follow sensor costs. The low-cost capacitive sensors were more accurate than manufacturer specifications, and could match the performance of the secondary standard sensor, after soil specific calibration. SMEC300 is accurate ( M A E , R M S E , and R A E of 2.12%, 2.88% and 0.28 respectively), precise, and performed well considering its price as well as multi-purpose sensing capabilities. The less-expensive SM100 sensor had a better accuracy ( M A E , R M S E , and R A E of 1.67%, 2.36% and 0.21 respectively) but poorer precision than the SMEC300. However, it was established as a robust, field ready, low-cost sensor due to its more consistent performance in soils (particularly the field soil) and superior performance in fluids. Both the capacitive sensors responded reasonably to variations in temperature and salinity conditions. Though the resistive sensors were less accurate and precise compared to the capacitive sensors, they performed well considering their cost category. The YL100 was more accurate ( M A E , R M S E , and R A E of 3.51%, 5.21% and 0.37 respectively) than YL69 ( M A E , R M S E , and R A E of 4.13%, 5.54%, and 0.41, respectively). However, YL69 outperformed YL100 in terms of precision, and response to temperature and salinity variations, to emerge as a more robust resistive sensor. These very low-cost sensors may be used in combination with more accurate sensors to better characterize the spatiotemporal variability of field scale soil moisture. The laboratory characterization conducted in this study is a prerequisite to estimate the effect of low- and very low-cost sensor measurements on the efficiency of soil moisture based irrigation scheduling systems.

Compound events of wet and dry extremes: Identification, variations, and risky patterns.

Compound hydrometeorological extremes have been widely examined under climate change, they have significant impacts on ecological and societal well-being. This study sheds light on a new category compound of contrasting extremes, namely compounding wet and dry extremes (CWDEs). The CWDEs are characterized as devastating dry events (EDs) accompanied by wet extremes (EWs) in a given time window. Notably, we first adopt a separate system to identify coinciding events considering the different evolving processes and impacting patterns of EDs and EWs. The peak-over-threshold and standardized index methods are used in a daily and monthly window to identify EWs and EDs respectively. Furthermore, the spatial-temporal changes and risky patterns of CWDEs are revealed by using the Mann-Kendall test, the Ordinary Least Squares, and the Global and Local Moran indices. Germany is the study case. As one major finding, the results indicate a pronounced seasonal effect and spatial clustering pattern of CWDEs. The summer is the most vulnerable period for CWDEs, and the spatial hotspots are mainly located in the southern tip of Germany, as well as in the vicinity of the capital city Berlin. Besides, robust uptrends in CWDEs across all evaluation metrics have been discovered in historical periods, and the moist climate and complex geography collectively contribute to severe CWDEs. Unexpectedly, the study finds that compounding events in dry regions are mainly driven by wet extremes, whereas they show a higher dependency on dry anomalies in wet regions. The research provides new insights into compound extremes which are composed of individual hazards with distinct features. Related findings will aid decision-makers in producing effective risk mitigation plans for prioritizing vulnerable regions. Lastly, the robust framework and open access data allow for extensive exploration of various compounding hazards in different regions.

Development and application of high resolution SPEI drought dataset for Central Asia.

Central Asia is a data scarce region, which makes it difficult to monitor and minimize the impacts of a drought. To address this challenge, in this study, a high-resolution (5 km) Standardized Precipitation Evaporation Index (SPEI-HR) drought dataset was developed for Central Asia with different time scales from 1981-2018, using Climate Hazards group InfraRed Precipitation with Station's (CHIRPS) precipitation and Global Land Evaporation Amsterdam Model's (GLEAM) potential evaporation (Ep) datasets. As indicated by the results, in general, over time and space, the SPEI-HR correlated well with SPEI values estimated from coarse-resolution Climate Research Unit (CRU) gridded time series dataset. The 6-month timescale SPEI-HR dataset displayed a good correlation of 0.66 with GLEAM root zone soil moisture (RSM) and a positive correlation of 0.26 with normalized difference vegetation index (NDVI) from Global Inventory Monitoring and Modelling System (GIMMS). After observing a clear agreement between SPEI-HR and drought indicators for the 2001 and 2008 drought events, an emerging hotspot analysis was conducted to identify drought prone districts and sub-basins.

A meta-analysis of the value of ecosystem services of floodplains for the Danube River Basin.

Floodplains provide ecosystem services (ES). Their evaluation is complex and integrative assessment remains challenging for sciences and practices. Studies have been published in the last two decades reporting ES monetary values of floodplains. Since ES are site-specific, we focus on those studies regarding the Europe's second largest river basin, namely the Danube River Basin (DRB). By analyzing these studies, we aim to answer the questions: "Do the significant predictor variables differ from previous meta-analyses?" and "Does the spatial database improve the meta-analysis?" In this context, we conducted a systematic review on Scopus and Web of Science combining the four themes "value", "ES", "floodplain", and "location". We conducted a meta-analysis of the Danube floodplains' ES values with different sub-groups based on the ES classes (provisioning, regulating, and cultural) and implemented model selection based on the corrected Akaike Information Criterion. We selected 251 entries from 25 studies to set up with a PostgreSQL spatial database, which provides limitless possibilities to enrich the information on the study areas. We observed that the most important variables to describe ES values of DRB floodplains depend on the ES class, but in general the area proportions of water bodies and riparian landscapes are important, together with the valuation method and the chemical or ecological status of the corresponding river section. Finally, we provided two versions of unconditional benefit-transfer functions to evaluate provisioning, regulating, and cultural ES. This paper complements previously conducted meta-analyses to recognize significant characteristics to value ES and it is a valid basis to help determine the ES value of Danube floodplains.

Fully automated snow depth measurements from time-lapse images applying a convolutional neural network.

Time-lapse cameras in combination with simple measuring rods can form a highly reliable low-cost sensor network monitoring snow depth in a high spatial and temporal resolution. Depending on the number of cameras and the temporal recording resolution, such a network produces large sets of image time series. In order to extract the snow depth time series from these collections of images in acceptable time, automated processing methods have to be applied. Besides classic image processing based on edge detection methods, there are nowadays ready-to-use convolutional neural network frameworks like Mask R-CNN that facilitate instance segmentation and thus allow for fully automated snow depth measurements from images using a detectable measuring rod. This study investigates the applicability of Mask R-CNN embedded in a newly developed work flow for snow depth measurements. The new method is compared to an automated image processing method carried out utilizing functionalities provided by the OpenCV library. The quality of both methods was assessed with the inclusion of manual evaluations of the image series. As a result, the newly introduced work flow outperforms the present classic image processing method in regard to stability, accuracy and portability. By applying the Mask R-CNN framework, the overall RMSE of two considered time series is reduced to approximately 20% of the value produced by means of the classic image processing approach. Moreover, the ratio of values within five centimeter deviation from the reference value was increased from 75% to 88% on average. Since no parameters have to be adjusted, the Mask R-CNN framework is able to detect known shapes reliably in almost any environment, making the presented method highly flexible.

Evaluation of homogenization methods for seasonal snow depth data in the Austrian Alps, 1930-2010.

Despite the importance of snow in alpine regions, little attention has been given to the homogenization of snow depth time series. Snow depth time series are generally characterized by high spatial heterogeneity and low correlation among the time series, and the homogenization thereof is therefore challenging. In this work, we present a comparison between two homogenization methods for mean seasonal snow depth time series available for Austria: the standard normal homogeneity test (SNHT) and HOMOP. The results of the two methods are generally in good agreement for high elevation sites. For low elevation sites, HOMOP often identifies suspicious breakpoints (that cannot be confirmed by metadata and only occur in relation to seasons with particularly low mean snow depth), while the SNHT classifies the time series as homogeneous. We therefore suggest applying both methods to verify the reliability of the detected breakpoints. The number of computed anomalies is more sensitive to inhomogeneities than trend analysis performed with the Mann-Kendall test. Nevertheless, the homogenized dataset shows an increased number of stations with negative snow depth trends and characterized by consecutive negative anomalies starting from the late 1980s and early 1990s, which was in agreement with the observations available for several stations in the Alps. In summary, homogenization of snow depth data is possible, relevant and should be carried out prior to performing climatological analysis.

Evaluation of precipitation input for SWAT modeling in Alpine catchment: A case study in the Adige river basin (Italy).

Precipitation is often the most important input data in hydrological models when simulating streamflow. The Soil and Water Assessment Tool (SWAT), a widely used hydrological model, only makes use of data from one precipitation gauge station that is nearest to the centroid of each subbasin, which is eventually corrected using the elevation band method. This leads in general to inaccurate representation of subbasin precipitation input data, particularly in catchments with complex topography. To investigate the impact of different precipitation inputs on the SWAT model simulations in Alpine catchments, 13years (1998-2010) of daily precipitation data from four datasets including OP (Observed precipitation), IDW (Inverse Distance Weighting data), CHIRPS (Climate Hazards Group InfraRed Precipitation with Station data) and TRMM (Tropical Rainfall Measuring Mission) has been considered. Both model performances (comparing simulated and measured streamflow data at the catchment outlet) as well as parameter and prediction uncertainties have been quantified. For all three subbasins, the use of elevation bands is fundamental to match the water budget. Streamflow predictions obtained using IDW inputs are better than those obtained using the other datasets in terms of both model performance and prediction uncertainty. Models using the CHIRPS product as input provide satisfactory streamflow estimation, suggesting that this satellite product can be applied to this data-scarce Alpine region. Comparing the performance of SWAT models using different precipitation datasets is therefore important in data-scarce regions. This study has shown that, precipitation is the main source of uncertainty, and different precipitation datasets in SWAT models lead to different best estimate ranges for the calibrated parameters. This has important implications for the interpretation of the simulated hydrological processes.

Evaluation of eight high spatial resolution gridded precipitation products in Adige Basin (Italy) at multiple temporal and spatial scales.

This study provides a comprehensive evaluation of eight high spatial resolution gridded precipitation products in Adige Basin located in Italy within 45-47.1°N. The Adige Basin is characterized by a complex topography, and independent ground data are available from a network of 101 rain gauges during 2000-2010. The eight products include the Version 7 TRMM (Tropical Rainfall Measuring Mission) Multi-satellite Precipitation Analysis 3B42 product, three products from CMORPH (the Climate Prediction Center MORPHing technique), i.e., CMORPH_RAW, CMORPH_CRT and CMORPH_BLD, PCDR (Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks-Climate Data Record), PGF (Global Meteorological Forcing Dataset for land surface modelling developed by Princeton University), CHIRPS (Climate Hazards Group InfraRed Precipitation with Station data) and GSMaP_MVK (Global Satellite Mapping of Precipitation project Moving Vector with Kalman-filter product). All eight products are evaluated against interpolated rain gauge data at the common 0.25° spatial resolution, and additional evaluations at native finer spatial resolution are conducted for CHIRPS (0.05°) and GSMaP_MVK (0.10°). Evaluation is performed at multiple temporal (daily, monthly and annual) and spatial scales (grid and watershed). Evaluation results show that in terms of overall statistical metrics the CHIRPS, TRMM and CMORPH_BLD comparably rank as the top three best performing products, while the PGF performs worst. All eight products underestimate and overestimate the occurrence frequency of daily precipitation for some intensity ranges. All products tend to show higher error in the winter months (December-February) when precipitation is low. Very slight difference can be observed in the evaluation metrics and aspects between at the aggregated 0.25° spatial resolution and at the native finer resolutions (0.05°) for CHIRPS and (0.10°) for GSMaP_MVK products. This study has implications for precipitation product development and the global view of the performance of various precipitation products, and provides valuable guidance when choosing alternative precipitation data for local community.