5 Changes in Water Management Strategies Over Time
KEY TOPICS
Assessments of groundwater microbial quality can strengthen future development of sensors. Dr. Helen Nguyen (University of Illinois at Urbana-Champaign) shared information about drinking water disease outbreaks during 2013-2014 that led to significant health effects. Approximately one-third of the outbreaks are related to groundwater, and one-third are related to surface water, and the final third is related to drinking water distribution systems. Most of the top causes of outbreaks in wells are microbial. A recent review of microbial pathogens in groundwater (Bradford and Harvey, 2017) includes information largely from the United States because it is the easiest data to obtain. Data from less developed countries would likely increase understanding significantly. Microbes, especially viruses, are not readily removed through sub-surface flow and microbes are highly mobile in the sub-surface environment. The specific characteristics of microbes prevent their attachment to the sand surface, allowing them to travel far within the sub-surface (Gutierrez and Nguyen, 2013; Liu et al., 2010). To understand water quality in rural environments, Dr. Nguyen discussed an example of the distribution of septic systems in the United States, which are usually used together with private wells (Lusk et al., 2017). Importantly, regions with high septic use are also prone to flooding (see Figure 5.1), and may also include infrastructure damage in areas where sampling occurs. The National Ground Water Association estimated that, in 2018, about 750,000 private wells were affected by hurricanes and flooding.1
Traditionally, the scope of remote sensing has been to determine changes in groundwater quantity, as pointed out by Dr. Nguyen. However, groundwater quality and the presence of microbial and chemical contaminants will be a serious challenge moving forward, especially given the uncertainty associated with extreme events and anthropogenic changes. It may be useful for future groundwater management efforts to monitor both quantity and quality, and a combination of in situ measurements, knowledge of contaminant fate and transport, together with modeling of groundwater integrated with surface water could be needed. She also shared thoughts on developing lidar or sensor network systems to aid management of groundwater over time, and a potential national or global database on pathogens, antibiotic resistance gene, genome sequencing, and links to hydrology. Understanding the connections between land use, civil and
___________________

agricultural infrastructure, and microbial quality of groundwater under extreme conditions will be crucial. A new institutional model for groundwater governance could help protect the quality as well as the quantity, she said.
Reflecting on the previous discussions, Dr. Michael Cosh (USDA) called participants attention back to the National Ground-Water Monitoring Network, specifically to consider how much more data may be needed, and how to ensure that newly collected data are “good” data in the context of decision-making. He pointed out that the United States is the most heavily monitored country in the International Soil Moisture Network, with an ongoing effort to create a database of the in situ soil moisture data in the United States. Some states have much more complete coverage than other states, which is the result of independently run state networks that must fund themselves, often by charging for data. With such a data collection model, contracts have to be negotiated to share data that can help inform decisions regarding drought estimation. Some of these issues can likely be solved through lessons learned from the groundwater community.
Dr. Cosh highlighted the cooperative observer program at the National Oceanic and Atmospheric Administration (NOAA) and the approximately 4,000 associated temperature equipment sites, made possible by citizens collecting and sharing data. This helps to create a climate record over time, and he asked participants to consider the possibility of scaling these
efforts up to approximately 10,000 sites with cooperators sending water use or individual well data. Such an approach could be an avenue to collect additional groundwater data in a manageable and cost-effective way. Although this kind of effort would not include the most highly calibrated equipment, the data collected still have the potential to help inform decisions.
As noted previously, irrigation is the main source of groundwater withdrawal in the United States, followed by public water supply. Dr. Cosh challenged attendees to consider efficiencies that can be pursued and conservation efforts that could be targeted. The locations of irrigated land change over time, typically out of places that are water-challenged and into places with more readily available water. Beyond the well-known crops that require irrigation, a large amount of turf grass is irrigated in the United States, and this use of water is not being tracked with agriculture use. Innovative ways of monitoring (e.g., tracking well data, cooperative programs, monitoring social media and changing food usage at restaurants, etc.) may be needed to help manage these water resources. Unexpected new sources of information may be available to use for research and decision-making purposes.
Following on the previous discussion on trends in water use, Dr. Sankar Arumugam (North Carolina State University) highlighted irrigation water use across the United States and reiterated the importance of understanding the human aspects of water use. He also discussed a USGS report that highlights data from water science centers in several states and illustrates key sources of uncertainty (acreage irrigated, application efficiency, and heterogeneity in irrigation methods) (Dickens et al., 2011). The report also provides guidance and recommendations to reduce the uncertainty in the data. Recent work compares surface water and groundwater irrigation withdrawals which indicate that surface water withdrawals in the West have been decreasing while groundwater withdrawals have generally increased (Das et al., 2018). This kind of information can provide useful information for water use management. No public data are available on area irrigated by surface water and groundwater separately. Irrigation efficiency has changed over time as well, partly due to increases in area irrigated under drip and sprinkler irrigation systems. Dr. Arumugam highlighted another example of rich data on water use in China, where water use data are collected each year. In the North China Plain, the withdrawal is significant due to pumping and groundwater extraction.
Dr. Arumugam noted that, in trying to incorporate human and socioeconomic aspects into the analysis, some level of surrogates and approximation will be needed to capture the key variables. Gaps in national water data include lack of continuous groundwater records; poor water quality information; infrequent water use and demand (only available every 5 years); lack of inflows, releases, and water supply plans for national dams; and lack of information on inter-basin transfer. Collection of data on all of the above aspects are necessary to understand groundwater as both surface water and groundwater are heavily interconnected. Data issues and associated uncertainties exist in all aspects, including, for example, water use and demand, remote sensing, in situ data, and precipitation in hilly regions. Similar uncertainty issues also exist in models as well, namely process representation and coupling. Uncertainties in data and modeling will exist in all forms of data, Dr. Arumugam said, but understanding how these uncertainties translate to uncertainties in decision making is also required. Data fusion (i.e., combining in situ, remote
sensing, and groundwater management data) may help reduce the uncertainty in decision-making, taking care to address the appropriate spatial and temporal scales.
DISCUSSION
Dr. Ali Akanda (University of Rhode Island) noted that researchers appear to monitor primarily in areas of extensive groundwater withdrawal. Dr. Cosh pointed out that the data are typically collected by aquifer, and so high densities of observations may exist for specific aquifers where funding was obtained to invest in wells or where a state-level interest exists for the data. Research methods occasionally have to be adapted to accommodate the data that are available.
Dr. Rajaram asked about in situ microbial sensors that are being developed. Dr. Nguyen noted that the methods used depend on the variable of interest. Genome sequencing, for example, can be a very powerful tool for tracking foodborne outbreaks and other public health issues, but could also provide useful information for groundwater. On the other hand, if the goal is to determine what type of bacteria or pathogen may be present, innovative new sensors are being developed to enable this work in the field.
Dr. Somani asked about the impacts of hydraulic fracturing on water quality as well as aquifer storage and recovery. Dr. Cosh explained that hydraulic fracturing can be thought of as a “highway in the soil.” It is a fracture that shortens residency times, and many times the ground is used as a filter. This process bypasses the natural ground filter. Dr. Rajaram added that most fracking is occurring at great depths that will likely not affect groundwater aquifers directly, but the disposal of water that is brought up after these operations needs to be handled near the surface in holding ponds which may pose some risk of contamination. Dr. Person replied that any contamination in Pennsylvania has been related to poorly installed wells, rather than the hydraulic fracturing itself.
Dr. Akanda wondered if the information on movement of rotavirus through groundwater is based on United States groundwater data, and if any studies exist in the developing world on this issue. Dr. Nguyen noted that rotavirus is located everywhere, but it strikes other countries much harder than the United States. The study that she mentioned earlier also revealed that rotavirus is able to mutate, and some animal rotavirus is much more persistent in the environment compared to human rotavirus. This can produce hybrid characteristics between the two, which is important to understand the zoonotic route of the infections.
Dr. Gou raised the issue of water use efficiency and water consumption, and noted that even with the same amount of water withdrawal, the consumption from various irrigation methods will be different and thus will change the impacts on the whole water cycle, including infiltration and ET. Traditional water resources management is focused mainly on water withdrawal rather than water consumption, but perhaps emerging studies will need to shift that focus to the impacts of the changes in irrigation method on water supply. Dr. Arumugam agreed that the community is starting to shift focus in this way, but will require access to robust water use data on long time scales.
In the final breakout session, participants discussed steps that may need to be taken to manage groundwater in areas of severe depletion. The impacts of groundwater depletion are
location-dependent, and may include migration in some cases. In general, better monitoring is needed, together with a focus on reducing water usage, reusing water, recycling wastewater, and other management techniques. Understanding human behavior and the context in which decisions are made can potentially help in avoiding inadvertent incentives to increase water usage (i.e., Jevons paradox2). Examples of case studies with high productivity and low water usage can shed some light on the benefits of focusing on technology and engineering solutions as well as seeking intelligent ways to recharge aquifers and look for high permeability geologic structures that connect to aquifers. In addition, rebuilding infrastructure in a sustainable way using integrated storm water management, decreasing impervious surfaces, and increasing infiltration can provide paths forward. One area of interest noted by participants was the feasibility of a database or way of mapping potential groundwater-related conflicts around the world. This may help increase understanding of the importance of this issue with the public and user community, especially if geopolitical water risk can be linked to everyday lives.
Remote sensing technologies and modeling capabilities may be useful to identify and track water management approaches. Participants highlighted the ability of InSAR to detect subsidence, but acknowledged that subsidence does not always have a consistent relationship with groundwater depletion (i.e., land subsidence can be caused by a number of other sources such as compaction, organic oxidation, or latent compaction). Optical and radar remote sensing methods can be used to understand recharge, and innovations like citizen science as well as cell phone applications can be used to track (and raise awareness of) water use and conservation. Patterns of change in soil moisture can also be used to infer information about water management approaches. For example, soil moisture will decline over the growing season, if irrigation occurs normally (avoiding too much irrigation). Other technologies may include high resolution thermal data, grid networks of sensors, high accuracy sensors, and bundling of sensors with pump manufacturers.
Participants discussed NGA resources in the context of water management strategies and noted that NGA generates data and serves as a repository for data. Data are occasionally purchased from commercial satellites, and some data sets are gathered from private sources but may have limitations for their use. NGA could be particularly helpful in translating science to non-specialists and policy-makers, and placing these issues in the broader context of the food-water-energy nexus. Some participants suggested that NGA may wish to consider producing educational resources, and together with other agencies, states, and local governments, think about water policies, water reuse activities, and investments that have already been made.
International agreements and collaborative work with international scientists to produce high quality data products could be an important future consideration. Similarly, cross-agency and transboundary efforts are desired as well. As an example of these collaborative scientific efforts, some participants suggested using a land use model based on SWOT to predict how land cover will affect groundwater recharge.
___________________
2 The Jevons paradox occurs when technological progress or government policy increases the efficiency with which a resource is used (reducing the amount necessary for any one use), but the rate of consumption of that resource rises due to increasing demand.
This page intentionally left blank.