Arid and semiarid ecosystems contain relatively high species diversity and are subject to intense use, in particular extensive cattle grazing, which has favored the expansion and encroachment of perennial thorny shrubs into the grasslands, thus decreasing the value of the rangeland. However, these environments have been shown to positively impact global carbon dynamics. Machine learning and remote sensing have enhanced our knowledge about carbon dynamics, but they need to be further developed and adapted to particular analysis. We measured the net ecosystem exchange (NEE) of C with the eddy covariance (EC) method and estimated gross primary production (GPP) in a thorny scrub at Bernal in Mexico. We tested the agreement between EC estimates and remotely sensed GPP estimates from the Moderate Resolution Imaging Spectroradiometer (MODIS), and also with two alternative modeling methods: ordinary-least-squares (OLS) regression and ensembles of machine learning algorithms (EMLs). The variables used as predictors were MODIS spectral bands, vegetation indices and products, and gridded environmental variables. The Bernal site was a carbon sink even though it was overgrazed, the average NEE during 15 months of 2017 and 2018 was −0.78 $\mathrm{g}\mathrm{C}{\mathrm{m}}^{-\mathrm{2}}{\mathrm{d}}^{-\mathrm{1}}$ , and the flux was negative or neutral during the measured months. The probability of agreement (θs) represented the agreement between observed and estimated values of GPP across the range of measurement. According to the mean value of θs, agreement was higher for the EML (0.6) followed by OLS (0.5) and then MODIS (0.24). This graphic metric was more informative than r² (0.98, 0.67, 0.58, respectively) to evaluate the model performance. This was particularly true for MODIS because the maximum θs of 4.3 was for measurements of 0.8 $\mathrm{g}\mathrm{C}{\mathrm{m}}^{-\mathrm{2}}{\mathrm{d}}^{-\mathrm{1}}$ and then decreased steadily below 1 θs for measurements above 6.5 $\mathrm{g}\mathrm{C}{\mathrm{m}}^{-\mathrm{2}}{\mathrm{d}}^{-\mathrm{1}}$ for this scrub vegetation. In the case of EML and OLS, the θs was stable across the range of measurement. We used an EML for the Ameriflux site US-SRM, which is similar in vegetation and climate, to predict GPP at Bernal, but θs was low (0.16), indicating the local specificity of this model. Although cacti were an important component of the vegetation, the nighttime flux was characterized by positive NEE, suggesting that the photosynthetic dark-cycle flux of cacti was lower than ecosystem respiration. The discrepancy between MODIS and EC GPP estimates stresses the need to understand the limitations of both methods.

Interception losses are a major influence in the water yield of vegetated areas. For most storms, rain interception results in less water reaching the ground. However, fog interception can increase the overall water storage capacity of the vegetation and once the storage is exceeded, fog drip is a common hydrological input. Fog interception is disregarded in water budgets of semiarid regions, but for some plant communities, it could be a mechanism offsetting evaporation losses. Tillandsia recurvata is a cosmopolitan epiphyte adapted to arid habitats where fog may be an important water source. Therefore, the interception storage capacity by T. recurvata was measured in controlled conditions and applying simulated rain or fog. Juvenile, vegetative specimens were used to determine the potential upperbound storage capacities. The storage capacity was proportional to dry weight mass. Interception storage capacity (C_min) was 0.19 and 0.56 mm for rainfall and fog respectively. The coefficients obtained in the laboratory were used together with biomass measurements for T. recurvata in a xeric scrub to calculate the depth of water intercepted by rain. T. recurvata contributed 20 % to the rain interception capacity of their shrub hosts: Acacia farnesiana and Prosopis laevigata and; also potentially intercepted 4.8 % of the annual rainfall. Nocturnal stomatic opening in T. recurvata is not only relevant for CO₂ but for water vapor, as suggested by the higher weight change of specimens wetted with fog for 1 h at dark in comparison to those wetted during daylight (543 ± 77 vs. 325 ± 56 mg, p = 0.048). The storage capacity of T. recurvata leaf surfaces could increase the amount of water available for evaporation, but as this species colonise montane forests, the effect could be negative on water recharge, because potential storage capacity is very high, in the laboratory experiments it took up to 12 h at a rate of 0.26 l h⁻¹ to reach saturation conditions when fog was applied.

Different components of sustainability have been measured in previous studies at national, regional, and local scales. However, few efforts have aimed to measure water sustainability at the basin level. In this work, a water management sustainability index (WMSI) for quantifying water sustainability at the sub-basin level in Mexico is proposed. Variables related to water management in the environmental, social, economic, and institutional sub-systems of each sub-basin system were selected. The variables were aggregated in four sub-indexes based on a factor analysis, the sub-indexes scores were codified to indicate the level reached of water sustainability. The WMSI classified the water sustainability of the sub-basins of Mexico between poor and deficient, and the code classification indicated the level of compliance of the 969 sub-basins proposed by the National Institute of Statistics and Geography with sustainability criteria.

Mexico is a diverse country in terms of culture and natural environments. For this reason, the delimitation of homogeneous basins with similar environmental, social, and economic attributes is important in order to facilitate the elaboration of high-impact regional development strategies. However, this represents an ongoing challenge due to the complexity of the interactions that occur within socio-ecological systems at a regional scale. In the present study, the main objective was to identify the interrelationships among different aspects of the socio-ecological system located within basins, with the goal of utilizing this information to promote the region-specific sustainable development of an Integrated Water Resources Management (IWRM). Therefore, in this study, environmental, social, economic, and institutional variables, relevant to water management and with the capacity to be expressed spatially, were utilized to identify regions with similar characteristics and to regionalize the urban sub-basins of Mexico based on a principal component analysis (PCA) and the k-medoids clustering algorithm. The identification of the most adequate number of regions at the national level was determined by the silhouette method. As a result, five distinct regions for Mexico were generated, which forms the first step in the design of integrated water resources management strategies for these regions.

Vegetation fixes C in its biomass through photosynthesis or might release it into the atmosphere through respiration. Measurements of these fluxes would help us understand ecosystem functioning. The eddy covariance technique (EC) is widely used to measure the net ecosystem exchange of C (NEE) which is the balance between gross primary production (GPP) and ecosystem respiration (R_eco). Orbital satellites such as MODIS can also provide estimates of GPP. In this study, we measured NEE with the EC in a scrub at Bernal in Mexico, and then partitioned into gross primary production (GPP-EC) and R_eco using the recent R package Reddyproc. Measurements of GPP-EC were related to the estimates from the MODIS satellite provided in product MOD17A2H, which contains data of the gross primary productivity (GPP-MODIS). The Bernal site was a carbon sink despite it was an overgrazed site, the average NEE during fifteen months of 2017 and 2018 was −0.78 g C m⁻² d⁻¹ and the flux was negative in all measured months. The GPP-MODIS underestimated the ground data when representing the relation with a Theil-Sen regression: GPP-EC = 1.866 + 1.861 GPP-MODIS; an ordinary less squares regression had similar coefficients and the R² was 0.6. Although cacti (CAM), legume shrubs (C₃) and herbs (C₃) had a similar vegetation index, the nighttime flux was characterized by positive NEE suggesting that the photosynthetic dark-cycle flux of cacti was lower than R_eco. The discrepancy among the GPP flux estimates stresses the need to understand the limitations of EC and remote sensors, while incorporating complementary monitoring and modelling schemes of nighttime R_eco, particularly in the presence of species with different photosynthetic cycles.

Medir la masa de forraje (MF) en la pradera, antes del pastoreo, es fundamental para determinar la asignación diaria de forraje en sistemas pastoriles de producción animal. La MF se estima por corte de forraje en áreas conocidas, utilizando ecuaciones alométricas, o con el uso de sensores de percepción remota (PR); sin embargo, la exactitud y practicidad de los distintos métodos para estimar la MF, es variable. El objetivo fue obtener modelos predictivos usando variables ambientales y del manejo de la pradera para predecir la MF. Se ajustaron modelos de regresión para estimar la MF con base en variables del manejo de la pradera (MP) o mediciones obtenidas por PR, como reflectancia, temperatura del aire y lluvia. Por tres años se estudió una pradera mixta pastoreada con bovinos productores de carne. Con 80 % de datos se modeló por mínimos cuadrados ordinarios (OLS) o por algoritmos de aprendizaje automatizado (ML). El 20 % restante de los datos se utilizó para validar los modelos usando el coeficiente de determinación y el sesgo promedio entre valores estimados y observados. El modelo base de estudio fue la relación entre la altura de la pradera antes del pastoreo y la MF de este modelo se ajustó usando OLS; la r2 fue 0.43. Cuando se ajustaron modelos que incluyeron variables del MP, la r2 fue 0.45 para OLS y 0.63 para ML. Al ajustar modelos con variables de MP y PR, la r2 fue 0.71 para OLS y 0.96 para ML. Los ensambles de modelos ajustados con ML redujeron el sesgo de estimados de MF de la pradera examinada. En general, los modelos de ML representaron mejor la relación entre altura de la pradera antes del pastoreo y MF que los de modelos de OLS, al ajustarlos con variables de manejo de la pradera y con información de PR. Los modelos de ML pueden usarse como herramienta para la toma de decisiones diaria en sistemas productivos pastoriles.