The Edwards Aquifer Authority (EAA) and other Permittees are at the beginning stages of implementing a complex Habitat Conservation Plan (HCP) and are doing an excellent job in many respects, such as fountain darter biomonitoring. Nevertheless, in the course of reviewing the EAA’s modeling and monitoring efforts for this report, the Committee has identified a number of overarching concerns regarding the implementation process—concerns that may hinder the later stages of this HCP and especially any future attempts to renew the HCP and the Edwards Aquifer Incidental Take Permit. This chapter presents these overarching concerns and offers suggestions for broader-based improvements to the HCP’s implementation. These suggestions all underscore a recommendation that the EAA and other Permittees begin thinking now about how best to manage the very long-term process of protecting the Endangered Species Act (ESA)-listed species, from data management and analysis to scenario planning.
The present suite of data collection activities and analyses found in the HCP combines monitoring, modeling, and various individual experiments and field studies into a larger overall science program. This multifaceted approach has advantages in being flexible and efficient because the various small pieces can be modified and staffed by relatively few people and new pieces can be added relatively quickly. However, without careful attention to integration and coordination across the entire project, inconsistencies in methods and analyses among the individual studies can occur, key elements
of study can be omitted, and observational data may not be collected in a manner that best informs model development and evaluation. Because the EAA contracts with outside groups to conduct much of the research and monitoring, and each of the contractors may not be fully aware of all program elements, it is particularly important that the EAA place special emphasis on careful integration of the overall program. Increased effort to integrate and synthesize data and research would enable the clear explanation of a cohesive set of results and conclusions that would increase the transparency and credibility of the science underlying the HCP.
Without clear attention to project integration, there is danger that the EAA efforts might result in a number of separate projects that do not combine seamlessly into an overall science program. For example, the water quality and biological monitoring programs developed somewhat independently and are not well integrated with each other (see Chapter 4), nor are they integrated with the hydrologic monitoring done outside the HCP (see Box 2-3). Sampling sites are not co-located to the extent that they could be. In addition, biological monitoring is largely focused on the various species of interest and could benefit from a broader focus on the biological communities in which these species are embedded and the multiple drivers that can influence these biological communities.
There also appears to be a lack of integration between the hydrogeologic modeling and research efforts. The hydrogeologic science investigations are conducted largely separately from the HCP and are not included in the Applied Research Program. While the hydrologic science program (Box 2-3) provides critical information for modeling and ecosystem assessment, the lack of a formal commitment means that research priorities may not be directed toward key questions in the HCP, especially if budgets become limited. For example, while there are plans to investigate the connection between the Comal and San Marcos pools under different hydrologic stresses, the priority of this research in the hydrogeologic science program is not clear. Nor is there a direct opportunity for groundwater modeling studies to inform the research plans. For example, as pointed out in the modeling sections of Chapter 2, specific tracer tests done by the hydrologic science program might benefit the conceptual model of the aquifer. Effects of climate change could be investigated by looking at long-term hydrologic data trends and incorporating them into the models, which would be helpful in developing adaptive management plans. Lack of coordination between the research efforts and the modeling team could delay improvements to the hydrologic modeling efforts. That being said, it is hoped that more formal coordination of the hydrogeologic research with the modeling efforts in the HCP could be implemented without adding layers of review that might slow the research process.
The Committee has identified several steps that can be taken to enhance integration of the HCP program.
- Develop an overall conceptual model of the system, including hydrological, climate, and biological community components. Such a model can guide the development of quantitative modeling of sub-components, identify gaps in understanding, and provide context for understanding the responses of particular species of interest. This overall conceptual model (or models) should be integrated with and cross-referenced to the more focused conceptual models on key species population dynamics. Such a set of conceptual models would be particularly valuable to understanding the multiple drivers of fountain darter population dynamics, including spring flow, climate variability, land use change, water quality, predation, and habitat.
- Develop a unified data/information management system so that data are easily available to all project teams. This recommendation is described in the next section of this chapter.
- Convene an annual science meeting to discuss results, discover gaps in understanding, and help plan future studies and monitoring activities. Such a meeting should include all project and contract scientists, other university and agency scientists who might be interested in becoming involved in future studies, and various stakeholder groups. These meetings can provide excellent forums to discuss results to date and provide transparency in identifying future research, monitoring, and modeling needs. The ability of the Science Committee to effectively engage and advise the Applied Research and other programs could be enhanced through this process.
DATA AND INFORMATION MANAGEMENT
The HCP is data-intensive, including hydrological, meteorological, chemical, and biological data collected at a variety of spatial and temporal frequencies and extents. Users and providers of HCP-relevant data include a diverse set of individuals and groups from academia, non-governmental organizations, commercial institutions, and municipal, state, and federal agencies. Rich sets of legacy data on multiple aspects of the Edwards Aquifer have been collected by numerous groups prior to the adoption of the HCP. Ongoing data collection as part of specific short-term studies or long-term monitoring is planned or under way as part of the HCP. The hydrological and ecological modeling that forms a core part of the HCP will produce large amounts of model output. Finally, all of these efforts will be repeated in 15 years when the Incidental Take Permit (ITP) needs to be renewed. The data emanating from these various activities need to be organized, quality assured, maintained, and curated. Furthermore, the data
must be accessible, discoverable, reviewable, and useable by individuals or groups both within and outside of the HCP set of stakeholders.
The Committee strongly recommends that the EAA develop a comprehensive information management plan as soon as possible. Such a plan would ensure both internal and external access to relevant data over both the short and long term, facilitate data analyses and syntheses across multiple data types and sources, buffer against the potential turnover of key personnel, and increase transparency and communication across stakeholders as the HCP is implemented and evaluated. In short, a well-planned and implemented information management system will make all aspects of the HCP more likely to succeed. The need for high quality data organization is evident when reviewing the multiple reports that can include data collected using different methodologies and approaches. There appears to be no attempt to collate these data sets in a way to provide rigorous statistical evaluation of long-term trends or interactions among ecological components. Further, there appears to be no clearly defined standard operating protocol for data sharing and management. The plan should include multiple aspects of information management such as:
- definition of data types and formats ranging from raw data to metadata; what types of data are available and how are they characterized and organized;
- explicit data management plan, from the method of collecting and initially transferring data from the field into digital form, to follow-up data flow consisting of (but not limited to) quality control, analysis, synthesis and dissemination;
- agreements about which data, and types, will be centrally housed and which will be distributed among individual stakeholders;
- maintenance of database integrity including quality assurance and short- and long-term curation, archival and data back-up plans;
- description of the data access and sharing policy;
- creation of an accessible environment for the retrieval of information;
- facilitation of linkages among diverse data sets;
- documentation of metadata for data interpretation and analysis; and
- analysis of information management staffing needs.
Developing and implementing a comprehensive plan is not trivial, and adequate time and resources should be made available. Full-time staffing by trained information managers will likely be required throughout the life of the project. Other complex, data-intensive projects such as the Long-Term Ecological Research Network, the Consortium of Universities for the
Advancement of Hydrological Sciences, Inc., and the Ecological Society of America have developed functional information management and data registry systems that might serve as models for the EAA.
PERFORMANCE MONITORING OF MINIMIZATION/MITIGATION MEASURES
The HCP includes multiple minimization and mitigation (M&M) measures related to habitat and vegetation restoration, including reduction of recreational impacts, removal of exotic species, bank stabilization, and planting of Texas wild rice. It is unclear how the EAA is quantifying either the M&M measures themselves or the responses of the various target species, making it difficult to determine whether the measures are effective. The Committee recommends that the M&M measures be monitored for their performance. They should also be explicitly integrated into the ecological modeling and Applied Research efforts, preferably in the form of experimental analyses with the intent of testing and ultimately improving the ecological models.
NEED FOR ADDITIONAL DATA ANALYSIS
The Committee has observed that much of the data found in documents supporting the HCP do not include error bars or other measures that demonstrate the variability of the data or the uncertainty of model predictions. More formal statistical analysis, such as the incorporation of variance into estimated means and other summary statistics, would give additional credibility to the scientific basis of the HCP process. There is significant opportunity for exploring the key field data sets and model results, both ecological and hydrological, with more advanced statistical methods than simple summary statistics and graphical plotting. Some of the recent final reports from the Applied Research Program used standard statistical methods as part of their analyses, and this should be the continued expectation for all HCP research. The Committee recommends that the EAA undertake more formal and rigorous statistical analyses of its laboratory and field data.
Such statistical (exploratory) analyses do not replace the summary statistics and plotting analyses being done now, which are appropriate, but rather act as complementary analyses. Examples of where statistical methods should be applied with the ecological monitoring data are the derivation of the habitat suitability functions and power analyses of future trend evaluations. To date, habitat suitability functions have been drawn by hand using expert opinion and the available field data. A statistical definition of current conditions and the collection of data appropriate to define
statistically significant future trends are critical to the evaluation of those trends. With the accumulation of more field data and laboratory results, a more statistically based fitting of the habitat suitability functions should be conducted. Modeling of habitat utilization has become a very active area of research (Guisan and Zimmermann, 2000; Guisan and Thuiller, 2005; Knudby et al., 2010; Feyrer et al., 2011), partly because of interest in predicting the responses to climate change and partly because of the development of robust statistical methods. Statistically based habitat suitability functions would then be compared to expert opinion functions to assess similarities and differences, and both could be used in analyses to bound predictions of habitat changes in response to spring flows.
The biomonitoring field data will be increasingly used for examining temporal trends in the indicator species and their correlation to environmental variables such as flow. This will be especially true for the fountain darter in the index reaches. The issue of how to interpret the biological significance of sudden drops in the density of the indicator species will arise, and can be addressed by exploring trend detection methods and power analyses with the presently available data. These analyses will provide information on the likelihood that different magnitude and duration changes in the indicator species densities are significant relative to past variation. The exact methods to use depend on the specific statistical methods selected for the trend and correlation analyses and also on the preliminary results of those analyses. However, the general philosophy of power analyses (whereby the data are simulated with different types of variance and then sampled and analyzed with the trend and correlation methods) is well accepted and can be implemented in most statistical software packages (e.g., R statistical programming language). There are also exploratory data analysis methods that should be investigated to identify less-than-obvious patterns in spatial time series data, such as indicator densities and flows measured at multiple locations. These methods and analyses should be explored to help focus future field and lab data collection that would further elucidate flow–fish relationships.
The application of statistical methods in hydrological analysis is primarily associated with quantitative uncertainty analysis of model predictions, discussed in detail in Chapter 2.
POSSIBLE SCENARIOS OF FUTURE CONCERN AND SCENARIO PLANNING
One notable aspect of the current implementation of the HCP is a prevailing assumption that relevant conditions—both legal and ecological—will remain relatively stable throughout the Plan’s 15-year implementation period and that the worst case drought conditions were those observed dur-
ing the drought of record in the mid-1950s. For example, EAA representatives indicated that they plan to consider climate change and its predicted impacts on the Aquifer (and on demands for Aquifer water) only in the next phase of implementation.
The Committee recommends that the entities implementing the HCP begin to think now about possible worst case scenarios and their potential implications for both modeling and HCP implementation. On the modeling side, the Permittees should consider whether the models currently being developed rest on ecological assumptions that could be altered by a changing climate and, if so, whether potential or predicted alterations can themselves be incorporated into the model. Another modeling issue may be whether the models being developed can alert the Permittees to potential discontinuities and ecological thresholds in the Edwards Aquifer system under different future climate scenarios.
On the implementation side, considering potential future changes now could allow the Permittees to develop contingency plans—ecological, political, or legal—for future “worst case” scenarios, building adaptability, flexibility, and resilience into the HCP’s execution. Scenario building is a widely recognized and approved strategy for adaptation planning (Hannah et al., 2002; Shaw et al., 2009) that a number of resource agencies are employing throughout the United States (e.g., NPS, 2013). Moreover, scenario building could also help the Permittees to identify, in advance, situations that might require amendments to the HCP itself or to the legal authorities of the various Permittees. The Committee suggests that at least six potential future “worst case” scenarios are worth considering. It will be important for the Permittees to have the expertise (e.g., risk assessment, social sciences) needed to address these scenarios.
Scenario #1: Increased Groundwater Pumping from Exempt/Unregulated Wells Undermines the HCP’s Minimum Flow Requirements
Under the Edwards Aquifer Authority Act, wells that produce less than 25,000 gallons per day of water are exempt from many of the Act’s requirements (EAAA § 1.33; EAA Rules §§ 702.1(70), 711.14, 711.20). There has been anecdotal evidence during the drought of 2014 that a number of landowners are drilling exempt wells within the Edwards Aquifer region (although the EAA has not provided the Committee with the cumulative number of exempt wells). If landowners are increasingly using this legal exemption, their cumulative pumping could undermine the HCP’s minimum flow requirements, especially during drought years, regardless of how well the Permittees implement the HCP’s measures. As a result, the Permittees may have to ask the Texas Legislature to retroactively close this loophole in the Edwards Aquifer Authority Act.
Scenario #2: Drought Conditions Exceed the Drought of Record
The entirety of the HCP is built on the premise that the drought of record will be the worst that drought ever gets for the Edwards Aquifer region. What if that premise is wrong? What happens, for example, if the Edwards Aquifer region experiences a worse drought than the drought of record that lasts for several more years? What happens if the Edwards Aquifer region experiences several droughts of record (or even near droughts of record) in succession? The current drought in Texas began in 2011 and persists across much of the state, including the San Antonio region where Stage 3 Critical Period Management is in effect.
It should be noted that tree-ring analyses suggest that decadal-scale droughts, including mega-droughts of 15 to 30 years duration, have occurred in Texas at least once every 100 years since the 1500s (Cleaveland et al., 2011). Moreover, the external review (EARIP, 2012, Appendix I) of the Hardy (2009) report felt that “the presumption that a drought will mirror a previous one in climatic intensity or with the same combination of factors [water quality parameters same as in 1950s, watershed conditions similar (e.g., impervious cover), connectivity constraints similar, biotic interactions similar (e.g., invasive species threats the same), demand for water and recreational demand the same, etc.], does not seem probable.” Thus, the 1950s drought likely does not represent the true worst-case scenario as a baseline for hydrological modeling.
Scenario #3: Mismatch between Conservation Triggers and Hydrologic Changes
Many of the triggering events for the HCP’s water conservation measures are tied to hydrologic conditions on a particular date each year. Have the HCP Permittees considered the risks to the system, to the ESA-listed species, and to implementation of the Plan itself from a “perfect storm” of bad timing of these key events? For example, what happens if low-flow conditions occur immediately after the October triggering date for water conservation measures and last well into the next year?
Scenario #4: Climate Change Impacts Become Significant Faster Than Expected
The impacts of climate change are already being felt in many parts of the country, and one of the most important lessons for both modeling and planning is to not assume stationarity in baseline ecological conditions such as precipitation, evapotranspiration, and hydrologic flow. Have the HCP Permittees developed monitoring plans that will alert them to changing or
potentially changing baseline conditions, models that incorporate changing baseline conditions, and contingency plans that will allow the HCP to remain viable if climate change impacts become significant within the period of the ITP?
Scenario #5: High Court Affirmation of the Bragg Constitutional Takings Decision
Bragg, as noted in Chapter 1, found that limiting the overlying landowners’ ability to pump water from the Edwards Aquifer could result in a constitutional taking of their property rights, requiring payment from the EAA. This decision could potentially undermine the economic and legal assumptions of the HCP, and the Permittees should consider the following issues:
- If Bragg is upheld on appeal, to how many groundwater permits is it likely to apply?
- If Bragg is upheld on appeal, will the EAA still be able, financially, to implement the HCP and the Act’s restrictions on groundwater pumping in the Edwards Aquifer?
- If Bragg is upheld on appeal and its implications undermine the EAA’s ability to implement the HCP, how do the Permittees plan to move forward? Will they abandon the HCP entirely, understanding that ESA Section 9 “take” liability may arise as a result? Do they expect the Texas Legislature to intervene in support of the HCP, either financially or legally? Might management of the Edwards Aquifer at that point be turned over to the relevant federal agencies, as the original 1993 federal court decision threatened? In short, do the HCP Permittees have a contingency plan in place to deal with Bragg if that decision is upheld and its application undermines implementation of the current HCP?
Scenario #6: Subjugation to Aransas National Wildlife Refuge ESA Issues
The Edwards Aquifer is directly connected to the Guadalupe and San Antonio Rivers, which in turn flow to San Antonio Bay and the Aransas National Wildlife Refuge, which in turn provides habitat to the ESA-listed whooping crane. According to the HCP, the whooping crane was not included for coverage in the Edwards Aquifer HCP because it is believed that: (1) factors affecting the crane and its habitat are not under the control of the EAA and partner applicants for the ITP; and (2) whooping cranes would not be affected adversely by the Covered Activities. In addition, the HCP states that the minimization and mitigation measures developed for the activities covered by the proposed permit should provide greater stabil-
ity in the flows emerging from the spring systems at Comal and San Marcos Springs and, therefore, are expected to provide a potential net benefit to the habitat conditions for the ecosystem used by the crane.
Nevertheless, in March 2013, in The Aransas Project v. Shaw, 930 F. Supp. 2d 716, 786-88 (S.D. Tex. 2013), the U.S. District Court for the Southern District of Texas concluded that the Texas Commission on Environmental Quality had effectuated a Section 9 “take” of ESA-listed whooping cranes as a result of state-permitted diversions of fresh water. While the U.S. Court of Appeals for the Fifth Circuit reversed the liability finding on proximate causation grounds in June 2014 [The Aransas Project v. Shaw, 756 F.3d 801, 816-23 (5th Cir. 2014)], the case nevertheless still made clear that water flows in the larger watershed are important to the continued survival of the cranes. Moreover, the plaintiffs in that case considered whether they should include the EAA and other upstream water users in the litigation. Although the plaintiffs ultimately decided not to seek defendants so far upstream, the decision nevertheless suggests that the Edwards Aquifer could become tied to a much larger ESA recovery process and plan like the one that has enveloped the Snake River in Idaho (NOAA Fisheries, 2014). Have the HCP Permittees considered the implications for the Edwards Aquifer if it becomes linked to the Aransas National Wildlife Refuge in ESA recovery planning? What might that mean for Edwards Aquifer management and the necessary modeling for the system? How might such a linkage affect the water conservation requirements and pumping limitations imposed on the Edwards Aquifer? Would the needs of the whooping cranes in effect drive most or all of the management of the Edwards Aquifer?
Cleaveland, M. K., T. H. Votteler, D. K. Stahle, R. C. Casteel, and J. L. Banner. 2011. Extended Chronology of Drought in South Central, Southeastern and West Texas. Texas Water Resources Institute Texas Water Journal 2(1):54-96.
EARIP. 2012. Habitat Conservation Plan. Edwards Aquifer Recovery Implementation Program.
Feyrer, F., K. Newman, M. Nobriga, and T. Sommer. 2011. Modeling the effects of future outflow on the abiotic habitat of an imperiled estuarine fish. Estuaries and Coasts 34:120-128.
Guisan, A., and W. Thuiller. 2005. Predicting species distribution: offering more than simple habitat models. Ecology Letters 8:993-1009.
Guisan, A., and N. E. Zimmermann. 2000. Predictive habitat distribution models in ecology. Ecological Modelling 135:147-186.
Hannah, L., G. F. Midgley, and D. Millar. 2002. Climate change-integrated conservation strategies. Global Ecology and Biogeography 11(6):485-495.
Hardy, T. B. 2009. Technical Assessments in Support of the Edwards Aquifer Science Committee “J Charge” Flow Regime Evaluation for the Comal and San Marcos River Systems. River Systems Institute, Texas State University.
Knudby, A., A. Brenning, and E. LeDrew. 2010. New approaches to modelling fish-habitat relationships. Ecological Modelling, 221:503-511.
NPS (National Park Service). 2013. Using Scenarios to Explore Climate Change: A Handbook. National Park Service Climate Change Response Program. Fort Collins, CO. Available at http://www.nps.gov/subjects/climatechange/upload/CCScenariosHandbookJuly2013.pdf.
NOAA Fisheries, West Coast Region. 2014. Recovering Snake River Sockeye Salmon. http://www.westcoast.fisheries.noaa.gov/publications/recovery_planning/salmon_steelhead/domains/interior_columbia/snake/snake_sockeye_proposed_recovery_plan_fs_071614.pdf.
Shaw, A., S. Sheppard, S. Burch, D. Flanders, A. Wiek, J. Carmichael, J. Robinson, and S. Cohen. 2009. Making local futures tangible—Synthesizing, downscaling, and visualizing climate change scenarios for participatory capacity building. Global Environmental Change 19(4):447-463.
