Within the monitoring cycle, both the bottom-up approach and top-down are incorporated. The bottom-up approach is in-situ data, measuring data at a single point, which can high a temporal resolution (minutes) but does not cover a large area and can be expensive. For example, a WMO complaint will range from 20,000 - to 50,000 USD , not including the annual maintenance cost, and the data may not be available to the public. On the other hand, Earth observation can cover a large area t but can either have a poor spatial resolution or temporal resolution that will not meet in-situ data. Still, the data will be available to the public.

Both data are used together since in-situ data is required to calibrate and validate satellite algorithms, allowing earth observation to fill in missing data gaps.

Hydroinformatics can help monitor or estimate all key aspects of the hydrologic cycle as relevant for a basin. These can include precipitation, snow/glacier, surface water levels (incl. in surface storage) and streamflow, groundwater levels, extraction, and recharge, evaporation from water bodies, evapotranspiration from vegetation, water quality, surface, and groundwater dynamics, forecasting, and other related parameters to give a comprehensive perspective of the water cycle.

Click on various parts of the interactive schematic below to explore illustrative ways in which a new world of in-situ and earth observation monitoring systems can help estimate these parts of the hydrologic cycle. Also, explore illustrative types of data visualizations with these data available generally at a global level.

An additional challenge is that in addition to national flows of the water cycle, it is important to look at how this supply is stored and used in a spatial context – such as a river basin or watershed. This requires a more comprehensive hydroinformatics+ view of not only the water aspects but also the land, vegetation, and other uses of water to better understand the inter-relationships in a spatial, hydrologic context as seen below.

Bottom-up (In-Situ)

There is an ever-evolving set of monitoring instrumentation to help estimate various parts of the hydrologic cycle and associated parameters (e.g., weather, surface and ground water levels and quality, soil moisture, etc.). Some of these (e.g., 3D-PAWS) also use innovative low-cost 3D printed housing to reduce costs in some cases. Data transfer is moving from legacy manual systems to dataloggers to now near real-time telemetry (ranging from using broadband to GSM to satellite depending on the situation). It is expected that billions of sensors will be integrated into many devices leading to an Internet of Things (IoT) – going from an estimated 8.74 billion devices in 2020 to over 25 billion IoT devices a decade later. Countries around the world are scaling up the use of such sensors in the public and private sector.

Installed Sensors

There are many innovations in sensor technologies that are resulting in lower cost and more reliable in-situ monitoring systems. The evolving interactive catalog below provides an interactive way to explore these – click on an icon to learn more about that technology (including advantages, disadvantages, and field photos where available).

Explore the eBook on Monitoring Systems here:

Leveraging 3D Printing: 3D-PAWS

Weather stations worldwide fail or are terminated because of improper location, poor maintenance, and lack of real-time communication. The University Corporation for Atmospheric Research (UCAR) and the US National Weather Service International Activities Office (NWS IAO), with financial support from USAID and OFDA, devised an affordable, easily repairable meteorological system for areas that have limited monitoring. The surface weather station will take about a week to build and will cost $200-400 USD, utilizing local materials, microsensors, a single-board computer, and a 3D printer. 3D-PAWS measures all major weather components, including air pressure and temperature, precipitation, relative humidity, and sunshine. Although 3D-PAWS does not fulfill WMO standards, the data generated can be incorporated into a decision support system to provide guidance on flash floods, disease outbreaks such as meningitis or malaria, agricultural monitoring, or regional weather forecasting. The system has been used in the US, Kenya, Uganda, Zambia, Barbados, and Curacao.

Sound monitoring

As water leaks are difficult to detect, Singapore PUB is deploying acoustic monitoring on their drinking water supply system to reduce non-revenue water. Hydroponic sensors record sounds within pipes and is sent to the cloud for analysis. Machine learning and predictive analytics are then used to identify the location of leaks and estimated size and recorded on a GIS map as an alert.

Stream gauges are now incorporating an Acoustic Doppler current profiler (ADCP) which measures velocity from the Doppler effect. A sound wave or ping is transmitted within the water, bounces off any sediment in the river, and returns to the instrument. Particles traveling away from the profiler will have a lower frequency, while incoming particles will be higher. The difference in frequency between the send and receive wave is called the Doppler shift. This shift determines particle and water velocity.

Data Rescue: Despite the rapid change in technology, many water agencies around the world still have many records in paper form or on many distributed databases such as spreadsheets on individual desktops. These could include information on well logs, fence diagrams, borehole records, old graphing rolls, streamflow records, water use, climate, metadata, computations, maps, etc. Wherever possible, there is a need to undertake “data rescue” – with the immediate assessment, categorization and digitization of relevant information, integration of these information into digital data in appropriate formats and ideally integrated into online services and a new online system to collect such information in the future. These can be done through many mechanisms, including as digitization service firms, interns, or in-house staff depending on volume of work guided by good quality management.

Portable Sensors

Surveys: The in-situ data from gauges can be augmented by mobile surveys that use portable sensors to undertake surveys of targeted areas – these can also be carried out on the ground, or in the water (including using water drones).

Online Surveys

There are many field apps using smartphones to collect geo-enabled data/multi-media (e.g. built on Open Data Kits or ODKs or proprietary services – e.g. KoboToolBox). Now, there are also versions that be done for online surveys, even with spatial content (e.g. working with experts to identify infrastructure rehab needs or with communities to outline extent of flooding or indicate potential areas to be monitored or visualization needs for a complex water system). These can be built using free, bundled, or subscription approaches (e.g. with KoboToolBox, ESRI Survey123 , etc.) to help collect data, convert them to services, and the visualize them through easy to create interactive dashboards to obtain insights. They could become part of the arsenal of a modern broad hydroinformatics system.

Crowdsourcing

There are a range of more digital methods that are starting to replace traditional log registers leveraging increasingly ubiquitous smartphone access (or SMS-based systems for legacy access). This facilitates better data entry and provides a platform for structured and unstructured crowdsourcing of data from targeted users and community monitoring and citizen report cards . These will become more prevalent as connectivity and mobile data access improves and, in a few years, 5G could also revolutionize edge computing and access speeds further.

The LOCSS (Lake Observation Citizen Science and Satellite) project aims to better understand lake dynamics by combining the efforts of citizen scientists with satellites. At 70 places worldwide, citizen scientists report lake height by reading lake gauge heights and then sending the data via text message or email. These results are combined with lake surface area data generated from satellite pictures. As a result, researchers can investigate how the volume of water in a particular lake fluctuates over time and assist in understanding how water travels in connection to these lakes and their surrounding terrain.

MyWell is a smartphone and SMS app for collecting and analyzing data related to depths to well water level, rainfall amounts, checkdam water levels and water quality parameters. The crowd sourcing app engages farmers, schools, and local communities, displays the status of the groundwater level in addition to the above parameters, with historical values for comparison and analysis.

TamilNadu water quality kit was developed to promote community participation and create awareness in schools. The kit uses reagents to test 12 water quality parameters, which include ph, hardness nitrate, phosphate, chloride, bacteriological examination, etc.

Extreme climatic events such as floods, droughts, and increasing sea levels make Indonesia susceptible to disasters. However , on the other hand, Indonesia has one of the greatest concentrations of Twitter users, where people post information on the weather, traffic jams, and school closings. PetaBencana is an open-source disaster platform that uses social media data to track disasters in real-time. Users are then sent to a website to identify a disaster on a map, geotag photographs, and submit a disaster report using a chatbot. The gathered social media data and official data are utilized to create a disaster map that makes vital information instantly available to everyone. The site can run on the most basic phones and poor data connectivity to increase overall accessibility.

PetaBencana uses a chatbot to encourage active user engagement in data verification and structure to cut costs and gain scale. The platform's success has been since the developers first learned how Indonesians convey information on social media. This technique is being applied abroad, but it does require an understanding of culturally sensitive behavior and risk perception so that the chatbot can be tailored to a new location, including an appropriate tone. There have been chatbot initiatives in other parts of the world, such as the Philippines.

Water Drones

There are a new generation of drones that are being deployed on and under water that bring a completely new dimension to hydroinformatics. For example, floating drones or sensors attached to floating craft or a Unmanned Underwater Vehicles (UUVs), some of which can be autonomous or tethered, can be used to survey temperature, bathymetry, water quality, plastics, dam safety, and even support fish assessments.

Top-Down (Earth Observation)

A whole new generation of new “eyes in the sky” has resulted in a range of new free (e.g., from NASA , ESA ) and subscription global services for earth observation that are not just using the visible spectrum but other wavelengths that can be used to see through clouds, determine vegetation health, estimate rainfall, evapotranspiration, and many other parameters that can be useful for water management.

Satellite Earth Observation

See an overview of how Earth Observation can be used in Water Resources Management. See an interactive e-book (Developed by NASA and the World Bank) summarizing some of the common water-related earth observation data services and additional discussion of options from the European Space Agency .

Climate Hazards Group InfraRed Precipitation with Station data (CHIRPS) is a quasi-global daily or pentad rainfall with 40-years of record starting from 1981- to the near present. CHIRPS blends satellite rainfall imagery with station data. A monthly video from 1981-2016 of Africa is shown in the video below.

NASA's Integrated Multi-Satellite Retrievals for Surface Precipitation, (IMERG) , provides most of the world's surface precipitation data. IMERG incorporates historical Tropical Rainfall Measurement Mission estimates from 2000 along with Global Precipitation Measurement. Multiple satellites allow for the development of half-hourly rainfall estimates at a 0.1° horizontal resolution.

Sentinel-1 is the first satellite in the European Space Agency's and the European Commission's Copernicus satellite constellation. The mission is made up of two satellites, Sentinel-1A and Sentinel-1B, that orbit in the same orbital plane. Sentinel-1 is equipped with a C-band Synthetic Imaging Radar (SAR) capable of imaging down to 5 meters and covering an area of up to 400 kilometers. SAR satellites are not affected by cloud cover or lack of illumination and can collect data over a day or night, under all weather conditions. It is planned to acquire high-resolution images of all global landmasses, coastal areas, and shipping channels, as well as vignettes of the global ocean.

The European Space Agency operates Sentinel-2 as part of the ESA Copernicus constellation. Twin optical high-resolution multispectral satellites flying in the same orbit but phased 180° apart are planned to have a 5-day revisit frequency at the Equator. Sentinel-2 carries an optical instrument payload that samples 13 bands where the spatial resolution ranges from 10 meters to 60 meters. Sentinel-2’s provides continuity of SPOT and Landsat-type image data by contributing to ongoing multispectral observations and are used for various purposes, including land management, agriculture and forestry, disaster monitoring, humanitarian relief operations, risk mapping, and security concerns.

Landsat 8 , launched in 2013, is the eighth satellite designed to continue the program of global space imagery acquisition that the United States has been implementing since 1972 with the help of the Landsat series. The multispectral satellite samples imagery in 11 bands: the visible, near-infrared, shortwave infrared, thermal and panchromatic band. The improved thermal band would improve the study of heat and moisture transfer in the agriculture and water management sector.

Earth Map is a collaboration of Google and FAO , it is a free, easy-to-use tool that accesses and analyses geospatial data. Earth map provides unprecedented access for visualizing and interacting with earth observation datasets and overcomes big data limitations for use in real-time monitoring and quick historical analyses.

Subscription Services: There is an increasing trend for private sector firms offering data from private sources (e.g., high-resolution satellite images) or analytics from proprietary algorithms on free data (e.g., using machine learning) as paid subscription services for users. This Data as a Service approach will become more popular to leverage cloud service and interoperability to allow users to mix a combination of free and such subscription services for their use in visualization or further analytics to obtain insights for decision support. These currently exist for a range of earth observation (e.g.Maxar,Airbus, Earth-i), climate, weather (e.g. Tomorrow, aWhere Planet Watchers), and water related services, and these will invariably grow rapidly in the future.

There are many publications that provide summaries of earth observation from a water perspective. A couple of these include an interactive e-book on Earth Observations for Water Resources developed with NASA by the Bank team.

In the area of irrigation Irriwatch provides detailed soil water potential and soil moisture data in the root zone using satellite remote sensing technology along with an estimate of actual crop evapotranspiration and crop production

Aerial Surveys

Over the last few years, researchers and development agencies have been using drones/planes to conduct aerial surveys and Heli-borne geophysical surveys of water resources that support vital ecosystems.

HeliBorne Geophysical Surveys: These surveys that used to be popular for oil exploration are increasingly being used for groundwater insights. One version (e.g., fromSkyTem) involves a large sensor ring suspended by a helicopter to provide insights to the geophysical structure 600m below the surface. These technologies can revolutionize the speed of collecting data over specific areas and the use of this information both for visualizing the invisible but also help with groundwater modeling.

Virtual Planet is a company that explores climate impacts through visualization based on hyper realistic models. They have leveraged the capabilities of drones, 3D mapping and virtual reality technology to recreate different coastal and inland hazards and help communities to learn about potential challenges and to adapt to a new climate. They use drones to get photos and specific information about the area that they are going to study. Then they combine that information with local elevation data, flood information and sea level rise projections to make site-specific renderings for what each place should expect to see with rising seas.

Eyes in the Sky, Pollution in the Water

Earth observation from satellites and Drones/UAVs are a novel approach to monitor water quality from space for some parameter (e.g., chlorophyll indicating algal blooms, sediment, cyanobacteria, etc.). For example, GEO AquaWatch is an initiative of the Group on Earth Observations (GEO) to build global capacity and utility of earth observation derived water quality data, products, and information to support water resources management and decision making. Many of these approaches leverage the rapid growth in availability of satellite information in the public domain and analytical tools such as Google Earth Engine. There also appears to be significant potential to use machine learning and other AI approaches to help fill gaps and estimate other water quality parameters to support decisions. A major challenge remains access to even basic consistent data services on water quality across the world.

Aerial Drones

Unmanned Aerial Vehicles (UAVs) or drones are becoming increasingly used as a part of the hydroinformatics toolkit. This includes use for just aerial video surveys, normal optical RGB sensor information (that can also be used to generate elevation information), and special sensors (e.g., LiDAR, multispectral, microwave, thermal, hyperspectral) to get additional insights. These are increasingly using cloud analytics (e.g. Pix4D, SiteScan ) to be able to do powerful online analytics and insights (see example ) using drone inputs to generate useful products for water resources planning and management (e.g., digital elevation models or digital terrain models to aid flood vulnerability assessments, land cover/NDVI or evapotranspiration analytics, etc.).