1. Introduction
A new critical zone monitoring network known as the Texas Water Observatory (TWO) has been developed to study how water, energy, carbon, and nutrients interact at multiple scales in Earth’s critical zone (CZ). CZ functionality is affected by land use and external meteorological forcing. For example, Houghton (2007) reported that the global land-use changes due to human activities alone have contributed over 156 PgC during the 150-yr span of 1850–2000. Likewise, a meta-analysis across different hydroclimates showed that the pulse size of a precipitation event regulates the carbon balance (Polley et al. 2010; Johnston et al. 2021). The variable infiltration depth impacts microbial respiration and plant gas exchange at different levels within the soil column (Huxman et al. 2004). Significant weather events, such as droughts, generally decrease gross primary production (GPP) and total ecosystem respiration (RECO) (Zscheischler et al. 2014). The impact of extreme events such as hurricanes, severe storms, floods, and droughts on these variables is poorly understood. Understanding the effects of land-use and climate change on CZ functionality is central to quantifying its vulnerability and sustainable management of various ecosystem services (e.g., production agriculture, grazing, forestry, coastal wetland, water supply, energy production, environmental health, and recreation). Furthermore, the effects of sequential disturbances (drought followed by flood and vice versa) on CZ functionality and resultant tipping points on CZ services (Field et al. 2012) are largely unknown. The deficit in knowledge raises important questions about how CZ processes respond to significant climate events such as droughts, floods, and hurricanes and how human alterations of the CZ evolve in space and time.
Human activities over time can degrade shallow and deep soils with important feedback to other components of the CZ (Foley et al. 2005). Few soil ecosystems are more anthropogenically altered than those in low-lying agricultural regions. Soil compaction and erosion reduce water-holding capacity, altering infiltration and runoff (Das Gupta et al. 2006; Alaoui et al. 2018) and eventually the amount of retained soil moisture. Under these conditions, soil loses organic matter, hindering microbial activities and reducing soil fertility. Reduced fertility then affects crop and forest yields (den Biggelaar et al. 2003; Montgomery 2007; Pimentel 2006), further impacting carbon sequestration (Adhikari and Hartemink 2016; Van Oost et al. 2012). In general, CZ soils under agricultural, peri-urban, forest, coastal marsh, or prairie land cover hold and release water, energy, carbon, and nutrients in a characteristic manner that includes “spatial hotspots” and “temporal hot moments” (Nelson and Boots 2008). The location of hotspots and timing of hot moments depend on heterogeneity, antecedent water and biogeochemical status, and the dominant environmental (natural or human induced) forcing or perturbation.
2. Gulf Coastal Plains: A natural laboratory to investigate climate and land-use change impacts
TWO is a river basin scale distributed facility that has been in operation since 2017 to monitor and investigate coupled water, energy, and carbon stores and fluxes at various space and time scales within the Gulf Coast Plains region. While the structure of the CZ in this dynamic region and its soil continuously evolve over decadal and longer time scales, their functionality is altered over scales of days to years by climatic disturbances (including extreme events) as well as various types of human modification (Banwart et al. 2019). Therefore, given the broad diversity of hydroclimates and CZ structures in the Gulf Coastal Plains, a key question that TWO aims to address is “What is the response of the CZ structure and its stores and fluxes to climate and land-use change?”
The Gulf Coast region in the south-central plains of Texas (extending from the coastline to 400 km inland; Fig. 1) is a unique natural laboratory characterized by meandering rivers [northwest–southeast (NW-SE)] oriented orthogonally to aquifer recharge zones [northeast–southwest (NE-SW)] and cascading watersheds with different land covers subjected to exceptional climatic variations within relatively short time and space windows. These characteristics result in rapidly evolving, scale-dependent hydrologic and biogeochemical processes in the critical zone. In addition to the natural variability of this region, vast areas of the Gulf Coastal Plains support intensive agriculture. The primary building blocks for this are the deep vertisols and vertic intergrade soils of the western Gulf Coast. These soils have high water- and nutrient-holding capacities (Coulombe et al. 1996; Wilding and Puentes 1988), but their surface properties have been heavily altered through cultivation and management. The shrink–swell nature of these soils adds complexity; their fine clay particles adsorb water and swell when wet (Braudeau and Mohtar 2006; Kim et al. 1999). In dry conditions, however, large deep cracks form, through which water readily enters and allows surface materials to fall into the cracks and be incorporated into lower horizons. After a certain time, surface materials mix into the subsoil, causing deep recycling of carbon and nitrogen stores (White 1985). Cultivation of these soils homogenizes the soil surface and interferes with the natural self-mulching behavior of the surface soil. Due to the mechanical turbation and structural variation found in these soils, fundamental soil hydrological processes such as infiltration, drainage, runoff, evapotranspiration, and root-zone soil moisture dynamics remain poorly understood.
Schematics reflect multiscale coupled biogeochemical and hydrologic processes in the CZ across the Gulf Coastal Plains of Texas, encompassing various land uses and land covers. As an example, the zoom-in view reflects the region’s vertisols, with swelling and shrinking clay accelerating bioturbation and coupled hydrologic–biogeochemical processes.
Citation: Journal of Hydrometeorology 25, 11; 10.1175/JHM-D-23-0201.1
The Gulf Coastal Plains region is also strategically important for understanding the impact of climate change on land use and management practices. During 2011–12, a severe drought affected large portions of the continental United States, including the Gulf Coast region. It was one of the most persistent droughts of its kind, partly because normal regional moisture recycling dynamics were impaired (Nielsen-Gammon 2012; Rupp et al. 2017).
On the contrary, Hurricane Harvey in 2017 delivered >1250 mm of rain within 2 days resulting in massive flooding and altered ecosystem functionality. More recently, in 2021, a historic snowstorm and ice storm in the region created unprecedented freeze and thaw conditions, further accelerating the dynamics in near-surface soil structure (e.g., Leuther and Schlüter 2021).
3. Texas Water Observatory: Network design criteria
TWO is located in south-central Texas along the Brazos River corridor with an outlet to the Gulf of Mexico. This network spans a mosaic of land uses (agriculture, range/pasture, native/managed prairie, forest, and coastal marsh) and landforms (low-relief erosional uplands to depositional lowlands). The observatory network crosses climatic and geologic gradients across a fetch of ∼400 km encompassing several watersheds and subwatersheds (Fig. 2). In essence, the TWO network of CZ observatories is located in five subwatersheds/watersheds and nine different land use/land covers across the Gulf Coastal Plains. The TWO network design is adapted to a low topographic relief region with gently sloping uplands, flat bottomlands, and deep fertile soils supporting various ecosystem services. The TWO network (Figs. 1 and 2) spans across several subwatersheds, addressing simultaneously the following geophysical and anthropogenic factors.
TWO network design spanning multiple subwatersheds/watersheds in the Brazos corridor, Texas (center plot). Green bullets show the primary TWO monitoring sites. Each primary site may comprise several nodes with different land covers (1—Riesel: aspirational agriculture, traditional agriculture, and native prairie; 2—Stiles: mixed prairie; 3—TAMU farm: improved pasture, maize row crop, and cotton row crop; 4—Danciger: hardwood forest; and 5—Sargent: salt marsh). Plots on the bottom show spatial maps for soils, major aquifers, land covers, ecoregions, and annual mean precipitation of Gulf Coastal Plains. Photos on the top show select primary sites with various instrumentation and land covers.
Citation: Journal of Hydrometeorology 25, 11; 10.1175/JHM-D-23-0201.1
a. Floodplains
CZ processes in low-relief, agricultural regions that sustain large, rapidly growing human populations are represented in the TWO network. Furthermore, deep floodplain soils contain paleoclimate and paleoenvironmental records that place the modern environment into a long-term geological context (Dearing et al. 2012; Sheldon and Tabor 2009). We concentrate our CZ network on processes occurring within floodplains and the processes that connect them to upland hillslopes. Sediments eroded from upland regions accumulate on lower-lying floodplains. Rapid aggradation on the floodplain limits soil development, whereas slow and/or intermittent aggradation allows deep soils to develop. Floodplain soils are eventually buried, providing excellent records of ancient critical zones (Aslan et al. 1998; Sheldon and Tabor 2009). The timing of aggradation and stabilization/pedogenesis during the late Quaternary on floodplains across central Texas is remarkably consistent. Buried soils that date to the specific time frames 7000–8000, 4500, 2500, 1000–600, and <500 radiocarbon years BP are common (Waters and Nordt 1995). The consistency suggests that floodplain dynamics during these periods were driven by external factors such as climate change rather than local geomorphic controls. Viewed in this context, the packages of alluvium and buried soils on central Texas floodplains comprise a rich archive to study how the CZ responds to external forcing over 100–1000-yr time scales. These floodplain records are most valuable when considered alongside data from modern monitoring studies as TWO. The synergy results in an opportunity to develop improved models of CZ evolution.
b. Fertile soils
Deep, clay-rich soils with high water- and nutrient-holding capacities (Kaiser et al. 2012; van der Meer 2000; Wattel-Koekkoek et al. 2001) sustain critical ecosystem services such as agricultural production for modern civilization. Alluvial deposits with shrink–swell soils are common throughout the Texas and Mississippi Embayment. The importance of these types of fertile soils is not limited to the Gulf Coast region. Large areas of the United States have residual soils underlain by Cretaceous-age mudrocks formed from smectitic clays; these soils occur across the “black belt” of Mississippi and Alabama and extend northward from Texas to North Dakota and southern Alberta. Thus, the TWO network design can help address the broader CZ science questions related to deep and clay-rich soils across the North American continent.
c. Subtropical climate
The subtropical climate is an important aspect of the TWO CZ monitoring network design. In the Gulf Coast region, rapid population growth is expected over the next few decades (Colby and Ortman 2015). It is necessary to investigate the sensitivity of subtropical CZ to anthropogenic activities. Climate norms in this region are punctuated by intense rain events and prolonged dry periods. Natural climatic drivers impact erosional and depositional environments over geological time scales; however, more recent anthropogenic disturbances (in the past 150 years after human settlement), including the cultivation of agricultural land, grazing, urban development, coastal development, and forestry, combine to make important, yet poorly understood, contributions to critical zone processes.
d. At the crossroad of gradients
The Gulf Coastal Plains cuts across several climatological, ecological, and geomorphological gradients, providing an excellent testbed for TWO network design. As shown in Fig. 2, the mean annual precipitation ranges from 1500 mm in the east to 230 mm in the west. Ecoregions are oriented predominantly parallel to the coast, geological formations, and aquifers gradient dip toward the coast, while river basins drain into the Gulf of Mexico.
e. Hydrologic extremes
The 2011–12 drought of record, the 2017 Harvey flooding, and the 2021 ice storm/snowstorm in the Gulf Coastal Plains warrant the development of strategies to manage and predict future drought, floods, or other extreme events. Long-term analysis of CZ processes using TWO data will significantly improve understanding of soil water dynamics, energy fluxes, and carbon fluxes in the critical zone under different land covers and landscapes during hydrologic extremes.
f. Urban centers
Central Texas has a unique combination of large, fast-growing population centers (Houston, Dallas, San Antonio, and Austin) along with a high risk of severe drought, flood, hurricanes, and other ecosystem service interruptions. TWO network design provides a good linkage for a better understanding of CZ processes across the rural to peri-urban corridor, leading to a better understanding and assessment of the risk and resilience of the region to certain natural calamities.
g. Land-use legacies
Since human settlement, anthropogenic impacts on the soil have appeared in stratigraphic columns at much higher resolution than those of natural processes (Autin and Holbrook 2012). The history of anthropogenic impact preserved in sedimentary deposits provides a valuable archive of past human impact on the environment (Dearing et al. 2012). Analysis of the recent stratigraphic record is necessary to understand the long-term impacts of anthropogenic activities on the evolution and functionality of Earth’s CZ. TWO provides a testbed to evaluate such impacts over the past 180 years. The geographically extensive land-cover change has a strong potential to alter biogeochemical processes and their linkages with the hydrologic cycle from ecosystem to regional spatial scales and across multiple time scales (O’Donnell and Caylor 2012; Scott et al. 2006).
4. Texas Water Observatory: Infrastructure
As described above, the selection of TWO network primary sites (Figs. 1 and 2) was based on their capacity to address fundamental CZ science, the availability of legacy data, and operational logistics. The TWO network comprises nine primary sites (Table 1), including several sites in collaboration with USDA Agricultural Research Service (USDA-ARS) Long-Term Agriculture Research (LTAR) at Riesel watersheds in Texas and U.S. Fish and Wildlife Services (USFWS) near San Bernard Wildlife Refuge on the Gulf Coastal margin. Located along the Gulf Coast Plains, the nine primary sites of TWO are distributed across four ecoclimatic regions: 1) Gulf Coast Prairies and Marshes, 2) Post Oak Savanna, 3) Blackland Prairie, and 4) South Texas Plains.
Location and geophysiography of TWO primary sites and collaborators. ASCL = Altoga silty clay loam, AFSL = Aris fine sandy loam, BC = Belk clay, BCC = Bacliff clay, CB = Churnabog clay, EL = Edna loam, HBC = Houston black clay, HC = Heiden clay, LL = Leton loam, PC = Pledger clay, RC = Rotex clay, SC = Surfside clay, WSCL = Westwood silty clay loam, WSL = Westwood silt loam, and YVFSL = Yahola very fine sandy loam. AmeriFlux site IDs of TWO sites, wherever applicable, are as follows: RFAA: US-Tx3, RFTA: US-Tx4, RFPR: US-Tx2, SFPr: US-Tx1, TFPr: US-Tx5, DAFo: US-Tx8, and SUSM: US-Tx9.
The Gulf Coast Prairies and Marshes region is characterized by its gently sloping terrain, with minimal elevation changes of less than 50 m, intersected by streams and rivers that flow into the Gulf of Mexico. It encompasses barrier islands along the coastline, marshy areas with salt grass surrounding bays and estuaries, remnants of tallgrass prairies, scattered oak parklands and oak mottes near the coast, and dense woodlands in the river valleys. Rainfall in this region ranges from 762 to 1270 mm annually, evenly distributed throughout the year. The growing season typically exceeds 300 days, marked by high humidity and warm temperatures. The soils are predominantly acidic sands and sandy loams, with clay deposits mainly found in the river bottoms. The indigenous flora primarily includes tallgrass prairies and live oak woodlands though brush species like mesquite and acacias have become more prevalent, especially in the western parts of the area. Pine woodlands are prominent in the eastern regions. Despite the appreciable loss of native habitat due to agriculture and urban development, the region remains vital for migratory bird populations and serves as a crucial spawning ground for various fish and shrimp species. The Post Oak Savanna region serves as a transitional zone, stretching northward toward the Great Plains or eastward into forested regions. Annual precipitation typically falls between 710 and 1015 mm, often peaking in May or June. Upland soils are typically light-colored and acidic sandy loams or sands, while bottomland soils range from light brown to dark gray and acidic, varying in texture from sandy loams to clays. The landscape features gentle rolling to hilly terrain, with elevations ranging from 100 to 270 m above mean sea level. This region can be characterized as Oak Savanna, where patches of oak woodlands dot grasslands. Cattle ranching is the primary agricultural activity in the Oak Savanna and Prairies, with introduced grasses like Bermuda grass being grazed alongside forage crops and native grasslands.
The Blackland Prairies region derives its name from the rich, fertile black soils that are characteristics of the region. Due to the fertility of the soil, much of the original prairie has been converted into agricultural land for cultivating food and forage crops. Annual rainfall averages between 710 and 1015 mm, with May being the peak month for rainfall in the northern part of the region, while the south-central area experiences a more evenly distributed rainfall throughout the year. Typically, the soils are dark-colored alkaline clays, often referred to as “black gumbo,” interspersed with some gray acidic sandy loams. The landscape is characterized by gentle rolling to nearly level terrain, with elevations ranging from 100 to 270 m above mean sea level. The primary agricultural activities in the region include crop production and cattle ranching.
The South Texas Plains extend from the outskirts of the Hill Country to the subtropical areas of the lower Rio Grande valley. Much of this region is arid, characterized by grasslands and thorny brush like mesquite and prickly pear cacti. The soils in the area range from alkaline to slightly acidic clays and clay loams. Deeper soils support tall brushes such as mesquite and spiny hackberry, while shallow caliche soils nurture short, dense brushes. Annual rainfall averages between 500 and 825 mm, with higher amounts observed moving from west to east. Rainfall is lowest in winter, peaking in spring (May or June) and fall (September). Summer temperatures are high, accompanied by exceptionally high evaporation rates.
Central to the TWO effort is the need to monitor, assimilate, model, and disseminate a variety of data such as climate/weather, soil moisture, soil temperature, matric potential, evapotranspiration, respiration, subsurface hydrogeophysics, groundwater, streamflow, ecosystem indicators, water quality, and water use/availability to understand the linked (blue and green) water, carbon, and energy cycle and provide various scenarios as tools for decision-makers. TWO was designed and implemented following continental observatory network protocols. Each primary site consists of one eddy covariance system [evapotranspiration (ET) node], spectral sensors and phenology camera (spectral node), three collocated soil-profiling and groundwater systems (soil node), and one Cosmic-Ray Soil Moisture Observing System (COSMOS node). The established infrastructure at the TWO network provides monitoring capability for near-real-time water, carbon, and energy fluxes in the critical zone. Distributed sensing of soil moisture profiles at different depths, energy balance components, carbon fluxes, groundwater level and quality, soil hydraulic properties, soil carbon status, primary production, weather parameters, and various land surface management and watershed attributes are regularly monitored using a nested design (Fig. 3a) including up to three soil nodes (accounting for the representative soil types) at each primary site (a representative site picture shown in Fig. 3b). TWO capacity includes more than 300 advanced field sensors (see Tables 2–4) deployed across all the primary sites under various land use/land covers, soil types, and climatic and geologic gradients along the Gulf Coastal Plains. A schematic of the TWO real-time data transfer system is shown in Fig. 4.
(a) TWO nested design for the primary site with ET node, soil nodes, spectral nodes, and COSMOS node for each subwatershed. (b) A typical observation node at a TWO network site (Riesel watershed) including eddy covariance, disdrometer, weather station, spectral sensors, phenocam, soil water content, potential, temperature sensors, and groundwater monitoring wells.
Citation: Journal of Hydrometeorology 25, 11; 10.1175/JHM-D-23-0201.1
TWO instrumentations at ET, spectral, and weather nodes at each primary site.
TWO soil water, groundwater, and surface water measurement systems at primary sites.
TWO supplemental water quality measurement systems.
TWO real-time data transfer system.
Citation: Journal of Hydrometeorology 25, 11; 10.1175/JHM-D-23-0201.1
a. ET node
The eddy covariance system at the ET node quantifies the exchange of heat, carbon dioxide, and water vapor between the Earth’s surface and the atmosphere. The system consists of an integrated 3D sonic anemometer with an infrared gas analyzer (IRGASON, Campbell Scientific) which is used to collocate vertical wind speed, carbon dioxide, and water vapor measurements. Energy balance is measured using a four-component net radiation (CNR4, Kipp and Zonen) sensor, two sets of soil pits with each instrumented with 12-cm water content reflectometer (CS655, Campbell Scientific) buried at 5-cm depth, averaging soil temperature probe (TCAV, Campbell Scientific) buried at 2- and 5-cm depths, and Hukseflux soil heat flux plate (HFP01) buried at 8-cm depth. The ET node at the Danciger site is supplemented by (heat dissipation) sap flow measurements in 20 trees within the tower footprint that are upscaled to transpiration.
b. Spectral node
The quantum sensors (LI-190, LI-COR) and spectral reflectance sensors (SRSs, Meter Environment) at the spectral node provide continuous monitoring of photosynthetically active radiation (PAR), normalized vegetation index (NDVI), and photochemical reflectance index (PRI) of plant canopies. PAR, NDVI, and PRI are used to monitor canopy biomass, leaf area, plant life cycle (green-up and senescence), biomass production, and light-use efficiency (LUE). The phenocam provides automated, near-surface remote sensing of canopy phenology across land-cover types. We use high-resolution digital security cameras (NetCam SC 5 MPIR Bundle) with infrared (IR) camera, 4.5–10-mm MP varifocal lens.
c. Soil node
At each soil node, profile moisture was measured at 5-, 15-, 30-, 75-, and 100-cm depths using water content reflectometers (CS655, Campbell Scientific). Matric potentials were measured at the same depths using MPS6 sensors (Meter Group, Pullman, Washington), matching the soil moisture measurements. In addition, CS 229-L heat dissipation sensors (Campbell Scientific) were also used to measure matric potential coincident with soil moisture. Nested groundwater wells (with varying depths based on geologic features of each primary site; Table 5) were drilled, and CTD-10 sensors (Meter Group) were used to measure groundwater level, electrical conductivity, and temperature. For the Sargent site, which is closer to the Gulf Coast, wells are outfitted with EXO1 sondes (YSI Inc.), which additionally measure pH, dissolved O2, CO2, and turbidity for a better understanding of tidal influences on groundwater.
TWO groundwater nested well depths.
Cosmic-Ray Soil Moisture Observing System is a newly developed method for measuring area-averaged soil moisture at the hectometer horizontal scale. COSMOS measures the neutrons that are generated by cosmic rays within the air, soil, and other materials; moderated by mainly hydrogen atoms located primarily in soil water; and emitted to the atmosphere where they mix instantaneously at a scale of hundreds of meters and whose density is inversely correlated with soil moisture.
d. Weather node
Each TWO primary site is supported by a weather node. Meteorological measurements are collected using a Vaisala temperature/relative humidity (RH) probe (HMP155A, Campbell Scientific) and two Texas Electronics rain gauges (TE525WS). In addition, laser precipitation disdrometers [OTT particle size velocity (Parsivel), OTT HydroMet, Germany] were deployed at our primary sites to capture the precipitation type, intensity, and drop size distribution.
e. Hydrogeophysics and soil hydraulic property campaigns
Periodic field campaigns to collect geophysical, soil hydrology, ecophysiological, proximal/remote sensing, and other supplemental data have been conducted at different TWO primary sites. Using soil cores collected at different depths by Geoprobe, conducting soil water retention and hydraulic conductivity measurements in the soil physics laboratory, and inverse modeling by HYDRUS, we populated the database for soil hydraulic and retention parameters.
In support of the TWO CZ mission, field-based geophysical equipment is available to map subsurface physical properties within the vadose and underlying saturated zone. Primary TWO geophysical infrastructure includes transient electromagnetic (G-TEM, Geonics Ltd.), ground-penetrating radar (GPR) Sensors and Software Inc. (GSSI), and electrical resistivity tomography (ERT; SuperStingAGI-USA) plus the associated processing and interpretation software. As examples of the capabilities, the G-TEM system has been used to map the freshwater–saltwater interface at the Sargent site, while the ERT system has been used to examine water uptake by trees at the Danciger site and groundwater–surface water interactions at the Texas A&M University (TAMU) farm site. Seismic, magnetics, and induced polarization geophysical equipment is also part of TWO infrastructure.
While TWO infrastructure is robust, certain intermittent challenges include 1) power supply and data transmission in remote locations, 2) soil sensor calibration in vertisols with swell–shrink soils, 3) regular sensor maintenance and calibration for maintaining data quality, 4) database design and access for push–pull upload, 5) sensor malfunction in extreme environments such as coastal margin with high salt, 6) sensor damage and relocation during crop growing seasons, and 7) wild animal damage. Addressing these challenges warrants regular maintenance and site-specific solutions.
f. Data management, quality assurance, and quality control
Data are collected and stored in dataloggers (CR6, CR1000X, and CR3000, Campbell Scientific) at the primary sites on a continuous basis. Most data are transmitted to a central cloud computing server [Elastic Compute Cloud (EC2) Windows**, Amazon Web Services (AWS)] via a cellular connection at scheduled intervals. From here, the raw data are uploaded to a relational database (MySQL**, AWS) where it is stored for investigator access (Fig. 4).
These data are next subjected to postprocessing and quality assurance/quality control (QA/QC; see discussion below), and the results are uploaded to a separate set of database tables. Based on the degree of gap filling and postprocessing, data are labeled as preprocessed and postprocessed. Preprocessed data refer to raw data from sensors. A web data portal (http://two.tamu.edu) makes select portions of the database accessible for public viewing. Real-time data from each field site can also be viewed using the EasyFlux Web platform (Campbell Scientific); this system is used primarily by the investigator team to track equipment status to ensure continued operations.
QA and QC are important to ensure TWO data quality. QA measures to reduce errors in flux and micrometeorological data include proper site selection and design, equipment choice and installation, data transmission and storage, timely sensor calibration, allowance for data user feedback, and regular maintenance scheduling. QC is implemented using both automated and manual visualization of variable ranges following Foken et al. (2004) and Pastorello et al. (2020). Gaps in flux data usually result from QA/QC procedures, missing data due to instrument failure, equipment calibration periods, and power losses. The REddyProc package (Wutzler et al. 2018) is used to gap fill flux data and partition CO2 fluxes into respiration and gross primary productivity. In addition, flux-data-qaqc (Volk et al. 2021) is used to postprocess gaps in energy fluxes through energy balance closure.
Figure 5 and Table 6 present examples of high-resolution (half-hourly and daily) (postprocessed) data as well as a monthly summary of different components of water, energy, and carbon cycles at different TWO primary sites. Carbon fluxes were partitioned into GPP and ecosystem respiration (RECO) from net ecosystem exchange (NEE). Positive NEE was defined as a net release of CO2 to the atmosphere. GPP was defined as RECO minus NEE. Positive GPP was defined as a positive uptake to the ecosystem. Figure 6 presents energy balance closure (EBC) at the TWO primary sites under different land covers, computed as the ratio (LE − H)/(Rn − G). The multiyear (2018–20) EBC ranged between 0.61 and 2.47. The EBC at the Sargent salt marsh (SUSM) site was exceptionally high because of the consistently flooded environment where the instruments were installed. In the rest of the sites, the closure problem is associated with low turbulent flows, which are often attributed to underestimation of latent or sensible heat fluxes or overestimation of net radiation or ground heat flux (Foken 2008).
Water, energy, and carbon observations for selected TWO observation sites with different land cover/use [i.e., (a) Stiles – pasture, (b) TAMU farm – agricultural land, (c) Danciger – forest]. (a-i),(b-i),(c-i) The energy observations include energy balance components at the land surface such as net radiation (Rn), latent heat flux (LE), sensible heat flux (H), and ground heat flux (G) (W m−2). The blue, red, green, and orange lines represent daily average values for Rn, LE, H, and G, respectively. (a-ii),(b-ii),(c-ii) The water observations include mean volumetric water content (VWC at a depth of 5 cm; m3 m−3), manual groundwater level depth (from the top of monitoring wells to the head of the water table; m), and the sum of daily precipitation (P) measurements (mm). (a-iii),(b-iii),(c-iii) Carbon observations include NEE, GPP, and RECO measurements. The red, green, and blue lines represent daily average values for the corresponding carbon-related observation (μmol CO2 m−2 s−1).
Citation: Journal of Hydrometeorology 25, 11; 10.1175/JHM-D-23-0201.1
Monthly summaries of environmental variables (2018–20) measured at different TWO sites. PCP: monthly total values (mm). VWC: average of profile VWC (m3 m−3) from triplicate soil nodes. Profile VWC is calculated as the arithmetic average of sensors at all (5, 15, 30, 75, and 100 cm) depths. LE, H, G, and Rn are in watts per square meter. NEE, GPP, and RECO are in μmol m−2 s−1.
EBC and the relationship between half-hourly turbulent fluxes (LE + H) and available energy averaged from 2018 to 2020. The darker blue gradient showcases the dense regions of data points.
Citation: Journal of Hydrometeorology 25, 11; 10.1175/JHM-D-23-0201.1
In conjunction with other remote sensing and legacy data, these high-resolution TWO datasets can help provide deeper insight for each of these balance components in different hydroclimates and land covers as diagnostic for coupled water, energy, and carbon budgets at different space time scales.
5. Outlook
TWO has been designed and developed with a long-term vision. The sustainability of growing human populations and the associated sustainability of diverse ecosystems in a changing environment are among the most pressing challenges for scientists, stakeholders, and decision-makers seeking to improve the livelihood of the world’s population. Variability in the regional weather and climate, including extremes (e.g., drought, flood, hurricane, and snowstorm) throughout the southeast United States, combined with increasing water and land demand, continues to have a significant social, political, economic, and environmental impact on the 50 million people who live in Texas and Gulf States. A greater understanding of the coupled water, energy, and carbon cycle and water availability, storage, usage, and quality, in both space and time, is critical for ensuring a sustainable future for the citizens of Gulf Coastal Plains. Central to addressing these challenges is the need to 1) develop a framework for guiding decisions based on accurate water, energy, and carbon cycle observations; 2) enable improved water management based on fluxes and stores to close the local, regional, and global water budget; 3) acquire, analyze, and provide value-added data products and indicators for water users on a sustainable and easily accessible manner; and 4) generate accurate regional hydrologic/biogeochemical/climatic modeling tools to better predict the short- and long-term water resources availability. TWO infrastructures will serve as a regional resource for better understanding and/or managing:
-
Agriculture: Drought monitoring, irrigation planning, root-zone soil moisture status, land and crop management, evapotranspiration dynamics, water–energy–food nexus, and water quality
-
Water resources: Water availability, streamflow, surface water storage, soil water storage, groundwater storage, water withdrawals, and water infrastructure planning
-
Ecosystems: Water and ecosystem services, flood and drought impacts, aquatic habitat in drought conditions, and wetlands and lakes
-
Biodiversity: Water stress and impacts in biota, water infrastructure and biodiversity, and gradients across urban and rural corridor
-
Disasters: Flood, drought, soil erosion and sedimentation, land degradation, and adaptation to climate variation
-
Health: Water quality, effect of flood and drought, and availability of potable water
-
Energy: Water for hydropower, cooling power plant, fracking shale gas, and biofuel production
-
Weather and climate: Improved regional modeling and forecast for precipitation, temperature, humidity, and extreme events
For example, Fig. 7a illustrates the carbon (C) assimilation characteristics of forest [Danciger forest (DAFo)] vs prairie [TAMU Farm prairie (TFPr)] ecosystems within the Gulf Coast, which can be broadly divided into three phases (phase 1: GPP ∼ RECO, phase 2: GPP > RECO, and phase 3: GPP < RECO). The annual C assimilation in DAFo is ∼1140.66 g C m−2 yr−1, which is 73.67% more annual assimilation compared to TFPr with ∼656.83 g C m−2 yr−1. These significant differences likely stem from various physiological and management factors. For instance, forests remain physiologically active throughout the year, whereas prairie’s carbon assimilation infrastructure is restricted to a narrower temporal window (Baldocchi and Penuelas 2019). Interestingly, the respiratory costs required to support carbon assimilation for both ecosystems are roughly of the same order (∼86% of annual GPP), which is likely driven by both energy and water availability. This is highlighted through absolute light-use efficiency and water-use efficiency (WUE) (Figs. 7b,c) which indicate an increase in both light and water usage during phase 2 of the C assimilation cycle. Although explorative in nature, these findings are indicative of the rich tapestry of hypothesis-driven questions that can be investigated using TWO data, such as:
-
Why do prairies (such as TFPr) have better light and water use efficiencies than forests (such as DAFo) during the summer season (Figs. 7b,c)?
-
What is the functional relationship between water, energy, and carbon cycle from local to regional scale across the ecosystem gradient in the Gulf Coast?
-
What is the sensitivity of the water, energy, and carbon budget to different hydrological fluxes and intermittent extremes at different scales?
-
Identify the scale-specific thresholds and their triggers (such as droughts, floods, and groundwater depletion) at which the water budget reaches a “tipping” point for different anthropogenic activities (such as agriculture and grazing) and evaluate recovery times.
-
How do local management decisions impact the regional water budget dynamics in TWO stations under different land use/land covers (LULC)?
-
What are the physical mechanisms that are responsible for the spatial and temporal variations in the strength of land–atmosphere coupling?
-
How can knowledge of land–atmosphere coupling be used to improve seasonal climate forecasting?
(a) Three-phase division of the carbon assimilation cycle for DAFo (forest) and TFPr (prairie) ecosystems. (b) LUE calculated as the ratio of cumulative GPP (g m−2 day−1) and Rn (W m−2) within each phase. (c) WUE calculated as the ratio of cumulative GPP (g m−2 day−1) and ET (W m−2) within each phase.
Citation: Journal of Hydrometeorology 25, 11; 10.1175/JHM-D-23-0201.1
Moreover, the emergent scaling relationship between coupled environmental processes remains severely unexplored. To name a few,
-
the complementary relationships between SM–ET, SM–soil respiration, and T–RECO;
-
vegetation and soil responses to precipitation pulse;
-
bidirectional streamflow–aquifer coupling;
-
development and propagation of droughts (impact of precipitation and temperature anomalies on hydrology and vegetation).
With seamless observations of the critical water, energy, and carbon cycle variables, TWO can serve as a testbed to explore these interlinkages between complementary environmental processes across a diverse selection of climate and land use.
Acknowledgments.
TWO observatory infrastructure was primarily funded by a Texas A&M University Research Development Fund (RDF) grant. Additionally, Texas A&M College of Agriculture and Life Sciences Water Chair and seed funding for the operation and maintenance of TWO are acknowledged. USDA-ARS funding support for Riesel sites is acknowledged. We also acknowledge the help of various students, technicians, and faculty during the project planning, field establishment, operation, repair and maintenance, and database development. BM developed the concept, secured resources, provided leadership for infrastructure development, and wrote the first draft of the manuscript. DM supported the field operation and data collection activities. GMi, GMo, ME, NR, CM, SC, and PK contributed to the planning of infrastructure development, facility/data management, and editing manuscript. NG, VS, AS, MH, DK, AD, SS, JMM, DM, RSi, BC, and RSo contributed to infrastructure development, operation, and maintenance and database development and management. DS and CJ were scientific collaborators for field operation.
Data availability statement.
This “vision” paper provides the design details and monthly summarized data at Texas Water Observatory for 3 years (2018–20) in Table 6. Daily average data (365 days across 2018–20) used for constructing Fig. 5 are uploaded to the Hydroshare database (http://www.hydroshare.org/resource/f8b42687e2db4ff89e468c6655f0611f). High-frequency raw data (with a 30-min sampling frequency) are available for collaboration only by requesting the project coordinator, Binayak Mohanty, at bmohanty@tamu.edu.
REFERENCES
Adhikari, K., and A. E. Hartemink, 2016: Linking soils to ecosystem services—A global review. Geoderma, 262, 101–111, https://doi.org/10.1016/j.geoderma.2015.08.009.
Alaoui, A., M. Rogger, S. Peth, and G. Blöschl, 2018: Does soil compaction increase floods? A review. J. Hydrol., 557, 631–642, https://doi.org/10.1016/j.jhydrol.2017.12.052.
Aslan, A., M. J. Kraus, and W. J. Autin, 1998: Holocene floodplain soils of the Mississippi River: Significance for the interpretation of alluvial paleosols. Quat. Int., 51–52, 36–37, https://doi.org/10.1016/S1040-6182(98)90198-7.
Autin, W. J., and J. M. Holbrook, 2012: Is the Anthropocene an issue of stratigraphy or pop culture? GSA Today, 22, 60–61, https://doi.org/10.1130/G153GW.1.
Baldocchi, D., and J. Penuelas, 2019: The physics and ecology of mining carbon dioxide from the atmosphere by ecosystems. Global Change Biol., 25, 1191–1197, https://doi.org/10.1111/gcb.14559.
Banwart, S. A., N. P. Nikolaidis, Y.-G. Zhu, C. L. Peacock, and D. L. Sparks, 2019: Soil functions: Connecting earth’s critical zone. Annu. Rev. Earth Planet. Sci., 47, 333–359, https://doi.org/10.1146/annurev-earth-063016-020544.
Braudeau, E., and R. H. Mohtar, 2006: Modeling the swelling curve for packed soil aggregates using the pedostructure concept. Soil Sci. Soc. Amer. J., 70, 494–502, https://doi.org/10.2136/sssaj2004.0211.
Colby, S. L., and J. M. Ortman, 2015: Projections of the size and composition of the U.S. population: 2014 to 2060. Population estimates and projections. US Census Bureau Current Population Rep. P25-1143, 13 pp.
Coulombe, C. E., L. P. Wilding, and J. B. Dixon, 1996: Overview of vertisols: Characteristics and impacts on society. Adv. Agron., 57, 289–375, https://doi.org/10.1016/S0065-2113(08)60927-X.
Das Gupta, S., B. P. Mohanty, and J. M. Köhne, 2006: Soil hydraulic conductivities and their spatial and temporal variations in a vertisol. Soil Sci. Soc. Amer. J., 70, 1872–1881, https://doi.org/10.2136/sssaj2006.0201.
Dearing, J. A., and Coauthors, 2012: Extending the timescale and range of ecosystem services through paleoenvironmental analyses, exemplified in the lower Yangtze basin. Proc. Natl. Acad. Sci. USA, 109, E1111–E1120, https://doi.org/10.1073/pnas.1118263109.
den Biggelaar, C., R. Lal, K. Wiebe, and V. Breneman, 2003: The global impact of soil erosion on productivity: I: Absolute and relative erosion-induced yield losses. Adv. Agron., 81, 1–48, https://doi.org/10.1016/S0065-2113(03)81001-5.
Field, C. B., V. Barros, T. F. Stocker, and Q. Dahe, Eds., 2012: Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation. Cambridge University Press, 582 pp., https://doi.org/10.1017/CBO9781139177245.
Foken, T., 2008: The energy balance closure problem: An overview. Ecol. Appl., 18, 1351–1367, https://doi.org/10.1890/06-0922.1.
Foken, T., M. Göockede, M. Mauder, L. Mahrt, B. Amiro, and W. Munger, 2004: Post-field data quality control. Handbook of Micrometeorology, X. Lee et al., Eds., Atmospheric and Oceanographic Sciences Library, Vol. 29, Springer, 181–208, https://doi.org/10.1007/1-4020-2265-4_9.
Foley, J. A., and Coauthors, 2005: Global consequences of land use. Science, 309, 570–574, https://doi.org/10.1126/science.1111772.
Houghton, R., 2007: Balancing the global carbon budget. Annu. Rev. Earth Planet. Sci., 35, 313–347, https://doi.org/10.1146/annurev.earth.35.031306.140057.
Huxman, T. E., and Coauthors, 2004: Precipitation pulses and carbon fluxes in semiarid and arid ecosystems. Oecologia, 141, 254–268, https://doi.org/10.1007/s00442-004-1682-4.
Johnston, A. S. A., and Coauthors, 2021: Temperature thresholds of ecosystem respiration at global scale. Nat. Ecol. Evol., 5, 487–494, https://doi.org/10.1038/s41559-021-01398-z.
Kaiser, M., R. H. Ellerbrock, M. Wulf, S. Dultz, C. Hierath, and M. Sommer, 2012: The influence of mineral characteristics on organic matter content, composition, and stability of topsoils under long-term arable and forest land use. J. Geophys. Res., 117, G02018, https://doi.org/10.1029/2011JG001712.
Kim, D. J., R. Angulo Jaramillo, M. Vauclin, J. Feyen, and S. I. Choi, 1999: Modeling of soil deformation and water flow in a swelling soil. Geoderma, 92, 217–238, https://doi.org/10.1016/S0016-7061(99)00033-6.
Leuther, F., and S. Schlüter, 2021: Impact of freeze–thaw cycles on soil structure and soil hydraulic properties. Soil, 7, 179–191, https://doi.org/10.5194/soil-7-179-2021.
Montgomery, D. R., 2007: Soil erosion and agricultural sustainability. Proc. Natl. Acad. Sci. USA, 104, 13 268–13 272, https://doi.org/10.1073/pnas.0611508104.
Nelson, T. A., and B. Boots, 2008: Detecting spatial hot spots in landscape ecology. Ecography, 31, 556–566, https://doi.org/10.1111/j.0906-7590.2008.05548.x.
Nielsen-Gammon, J. W., 2012: The 2011 Texas drought. Tex. Water J., 3, 59–95, https://doi.org/10.21423/twj.v3i1.6463.
O’Donnell, F. C., and K. K. Caylor, 2012: A model-based evaluation of woody plant encroachment effects on coupled carbon and water cycles. J. Geophys. Res., 117, G02012, https://doi.org/10.1029/2011JG001899.
Pastorello, G., and Coauthors, 2020: The FLUXNET2015 dataset and the ONEFlux processing pipeline for eddy covariance data. Sci. Data, 7, 225, https://doi.org/10.1038/s41597-020-0534-3.
Pimentel, D., 2006: Soil erosion: A food and environmental threat. Environ. Dev. Sustainability, 8, 119–137, https://doi.org/10.1007/s10668-005-1262-8.
Polley, H. W., and Coauthors, 2010: Precipitation regulates the response of net ecosystem CO2 exchange to environmental variation on United States rangelands. Rangeland Ecol. Manage., 63, 176–186, https://doi.org/10.2111/REM-D-09-00015.1.
Rupp, D. E., P. W. Mote, N. Massey, C. J. Rye, R. Jones, and M. R. Allen, 2017: Did human influence on climate make the 2011 Texas drought more probable? Bull. Amer. Meteor. Soc., 93, 1052–1054.
Scott, R. L., T. E. Huxman, D. G. Williams, and D. C. Goodrich, 2006: Ecohydrological impacts of woody-plant encroachment: Seasonal patterns of water and carbon dioxide exchange within a semiarid riparian environment. Global Change Biol., 12, 311–324, https://doi.org/10.1111/j.1365-2486.2005.01093.x.
Sheldon, N. D., and N. J. Tabor, 2009: Quantitative paleoenvironmental and paleoclimatic reconstruction using paleosols. Earth-Sci. Rev., 95 (1–2), 1–52, https://doi.org/10.1016/j.earscirev.2009.03.004.
van der Meer, J. J. M., 2000: The Nature and Properties of Soils. Sedimentology, 47, 1230–1230, https://doi.org/10.1046/j.1365-3091.2000.0376c.x.
Van Oost, K., G. Verstraeten, S. Doetterl, B. Notebaert, F. Wiaux, N. Broothaerts, and J. Six, 2012: Legacy of human-induced C erosion and burial on soil–atmosphere C exchange. Proc. Natl. Acad. Sci. USA, 109, 19 492–19 497, https://doi.org/10.1073/pnas.1211162109.
Volk, J., J. Huntington, R. Allen, F. Melton, M. Anderson, and A. Kilic, 2021: flux-data-qaqc: A python package for energy balance closure and post-processing of eddy flux data. J. Open Source Software, 6, 3418, https://doi.org/10.21105/joss.03418.
Waters, M. R., and L. C. Nordt, 1995: Late quaternary floodplain history of the Brazos River in East-Central Texas. Quat. Res., 43, 311–319, https://doi.org/10.1006/qres.1995.1037.
Wattel-Koekkoek, E. J. W., P. P. L. van Genuchten, P. Buurman, and B. van Lagen, 2001: Amount and composition of clay-associated soil organic matter in a range of kaolinitic and smectitic soils. Geoderma, 99, 27–49, https://doi.org/10.1016/S0016-7061(00)00062-8.
White, R. E., 1985: The influence of macropores on the transport of dissolved and suspended matter through soil. Advances in Soil Science, R. Lal and B. A. Stewart, Eds., Advances in Soil Science, Vol. 3, Springer, 95–120, https://doi.org/10.1007/978-1-4612-5090-6_3.
Wilding, L. P., and D. Tessier, 1988: Genesis of vertisols: Shrink-swell phenomena. Vertisols: Their Distribution, Properties, Classification and Management, L. P. Wilding and R. Puentes, Eds., Texas A&M Printing Center, 55–79.
Wutzler, T., A. Lucas-Moffat, M. Migliavacca, J. Knauer, K. Sickel, L. Šigut, O. Menzer, and M. Reichstein, 2018: Basic and extensible post-processing of eddy covariance flux data with REddyProc. Biogeosciences, 15, 5015–5030, https://doi.org/10.5194/bg-15-5015-2018.
Zscheischler, J., and Coauthors, 2014: Impact of large-scale climate extremes on biospheric carbon fluxes: An intercomparison based on MsTMIP data. Global Biogeochem. Cycles, 28, 585–600, https://doi.org/10.1002/2014GB004826.
