Abundant shallow underground brackish water resources could help in alleviating the shortage of fresh water resources and the crisis concerning agricultural water resources in the North China Plain. Improper brackish water irrigation will increase soil salinity and decrease the final yield due to salt stress affecting the crops. Therefore, it is urgent to develop a practical and low-cost method to monitor the soil salinity of brackish irrigation systems. Remotely sensed spectral vegetation indices (SVIs) of crops are promising proxies for indicating the salinity of the surface soil layer. However, there is still a challenge concerning quantitatively correlating SVIs with the salinity of deeper soil layers, in which crop roots are mainly distributed. In this study, a field experiment was conducted to investigate the relationship between SVIs and salinity measurements at four soil depths within six winter wheat plots irrigated using three salinity levels at the Yucheng Comprehensive Experimental Station of the Chinese Academy of Sciences during 2017–2019. The hyperspectral reflectance was measured during the grain-filling stage of winter wheat, since it is more sensitive to soil salinity during this period. The SVIs derived from the observed hyperspectral data of winter wheat were compared with the salinity at four soil depths. The results showed that the optimized SVIs, involving soil salt-sensitive blue, red-edge, and near-infrared wavebands, performed better when retrieving the soil salinity (R² ≥ 0.58, root mean square error (RMSE) ≤ 0.62 g/L), especially at the 30-cm depth (R² = 0.81, RMSE = 0.36 g/L). For practical applications, linear or quadratic models based on the screened SVIs in the form of normalized differential vegetation indices (NDVIs) could be used to retrieve soil salinity (R² ≥ 0.63, RMSE ≤ 0.62 g/L) at all soil depths and then diagnose salt stress in winter wheat. This could provide a practical technique for evaluating regional brackish water irrigation systems.

Leaf area index (LAI) is a key biophysical parameter for monitoring crop growth status, predicting crop yield, and quantifying crop variability in agronomic applications. Mapping the LAI at the field scale using multispectral cameras onboard unmanned aerial vehicles (UAVs) is a promising precision-agriculture application with specific requirements: The LAI retrieval method should be (1) robust so that crop LAI can be estimated with similar accuracy and (2) easy to use so that it can be applied to the adjustment of field management practices. In this study, three UAV remote-sensing missions (UAVs with Micasense RedEdge-M and Cubert S185 cameras) were carried out over six experimental plots from 2018 to 2019 to investigate the performance of reflectance-based lookup tables (LUTs) and vegetation index (VI)-based LUTs generated from the PROSAIL model for wheat LAI retrieval. The effects of the central wavelengths and bandwidths for the VI calculations on the LAI retrieval were further examined. We found that the VI-LUT strategy was more robust and accurate than the reflectance-LUT strategy. The differences in the LAI retrieval accuracy among the four VI-LUTs were small, although the improved modified chlorophyll absorption ratio index-lookup table (MCARI2-LUT) and normalized difference vegetation index-lookup table (NDVI-LUT) performed slightly better. We also found that both of the central wavelengths and bandwidths of the VIs had effects on the LAI retrieval. The VI-LUTs with optimized central wavelengths (red = 612 nm, near-infrared (NIR) = 756 nm) and narrow bandwidths (~4 nm) improved the wheat LAI retrieval accuracy (R² ≥ 0.75). The results of this study provide an alternative method for retrieving crop LAI, which is robust and easy use for precision-agriculture applications and may be helpful for designing UAV multispectral cameras for agricultural monitoring.

Carbon uptake of vegetation is controlled by phenology and photosynthetic carbon uptake capacity. However, our knowledge of the seasonal responses of vegetation productivity to phenological and physiological changes in alpine ecosystems is still weak. In this study, we quantified the spatio-temporal variations of vegetation phenology and gross primary productivity (GPP) across the source region of the Yellow River (SRYR) by analyzing MODIS-derived vegetation phenology and GPP from 2001 to 2019, and explored how vegetation phenology and maximum carbon uptake capacity (GPP_max) affected seasonal GPP over the region. Our results showed that the SRYR experienced significantly advanced trends (p < 0.05) for both start (SOS) and peak (POS) of the growing season from 2001 to 2019. Spring GPP (GPP_spr) had a significantly increasing trend (p < 0.01), and the earlier SOS had obvious positive effects on GPP_spr. Summer GPP (GPP_sum) was significantly and negatively correlated to POS (p < 0.05). In addition, GPP_max had a significant and positive correlation with GPP_sum and GPP_ann (p < 0.01), respectively. It was found that an earlier spring-summer phenology and higher photosynthetic peak enhanced the photosynthetic efficiency of vegetation in spring and summer and altered the seasonal patterns of vegetation productivity in the SRYR under warming and wetting climates. This study indicated that not only spring and autumn phenology but also summer phenology and maximum carbon uptake capacity should be regarded as crucial indicators regulating the carbon uptake process in alpine ecosystems. This research provides important information about how changes in phenology affect vegetation productivity in alpine ecosystems under global climate warming.

Testing reliability of the saturated soil hydraulic conductivity, K_s, estimated by applying the steady-state single-ring (SR) model to the quasi steady-state infiltration rates obtained with a single-ring pressure infiltrometer (PI) increases confidence in the estimated K_s values. Determining a means to estimate K_s from infiltration data collected with a square infiltrometer allows the use of sources of different shapes. Using numerically simulated infiltration rates for six homogeneous soils ranging in texture from sand to silty clay loam, this investigation suggested an overall good performance of the SR model, with estimated K_s values differing by not more than 25% from the true values for the 90% of the 96 considered runs. Larger errors were generally obtained for the silty clay loam soil. Even in this case, however, a small ring radius (0.038 m), a relatively high initial soil water content (initial effective saturation = 0.4) and a relatively high depth of ponding (0.10 m) allowed the obtainment of accurate predictions of K_s (error = 13%) with a run of practically sustainable duration (4 h). The SR model was also usable to analyze quasi steady-state infiltration data collected with a square infiltrometer when infiltration was assumed to occur through a circular source having the same area of the square infiltrometer. With this assumption, the estimates of K_s differed from the true values by not more than a practically negligible 16%. The results of this investigation should help better interpret K_s values obtained with the PI and also improve the experimental methodology, depending on the soil. Moreover, a wider applicability of the infiltrometer techniques, i.e. not limited to a circular source, can be expected. Soil heterogeneity should be taken into account in the future since heterogeneity is common in the field.

Core Ideas: The accuracy of the PI method in estimating soil K _s was tested by numerical simulations. Estimated K _s using two‐ponding‐depth and multiple‐ponding‐depth infiltration were compared. Transient and steady‐state infiltration data for six soils were used to estimate K _s . The PI should yield more accurate K _s estimates in coarse‐ than fine‐textured soils. The transient method does not solve the K _s inaccuracy problems in fine‐textured soils. The single‐ring pressure infiltrometer (PI) method is widely used to determine saturated soil hydraulic conductivity, K _s , directly in the field. The original and still most common way to analyze the data makes use of the steady‐state model developed by the Canadian School in the 90s and two (two‐ponding‐depth, TPD, approach) or more (multiple‐ponding‐depth, MPD, approach) depths of ponding. The so‐called Wu method based on a generalized infiltration equation allows analysis of the transient infiltration data collected by establishing a single ponding depth of water on the infiltration surface. This investigation, making use of simulated infiltration runs for initially unsaturated sand to silty clay loam soils, showed that, with a run duration of practical interest (e.g., 2 h), the PI can be expected to yield more accurate estimates of K _s in coarse‐textured soils than in fine‐textured soils even if the transient method is used instead of the steady‐state method. Performing a three‐level experiment and analyzing the estimated steady‐state infiltration rates with both the TPD and MPD approaches is a way to predict the reliability of the estimated K _s value. The K _s accuracy should be acceptable if the two approaches yield similar results. Otherwise, the MPD approach should be expected to yield more accurate K _s estimates than the TPD approach. The transient method does not solve the K _s inaccuracy problems in fine‐textured soils because obtaining accurate K _s data requires that the portion of total infiltration varying linearly with time represent a high percentage of total infiltration, but this percentage is small in fine‐textured soils when the run does not exceed a few hours. This investigation opens some new perspective on the use of infiltration data to make predictions on the expected reliability of the K _s calculations with reference to both steady‐state and transient data analysis procedures.

The Beerkan infiltration experiment is carried out by inserting the ring a short depth into the soil and establishing a positive head of water on the infiltration surface for at least a part of the run. Nevertheless, the data are analyzed by assuming a fully unconfined infiltration process (ring insertion depth, d = 0 cm) and a null ponded depth of water ( H = 0 cm). The influence of ring insertion and ponded water on an infiltration process of 2 h sampled every minute was tested in this numerical investigation. Five soils varying from sand to silt loam, three ring radii (5–15 cm), and the Beerkan‐specific range of values for both d and H (between 0 and 1 cm) were considered. The differences between the theoretical ( d = H = 0 cm) and the practical ( d = H = 1 cm) setups varied from −10.4 to +8.6% for the mean infiltration rate and from −10.2 to +8.3% for the final cumulative infiltration. These differences were small, and they decreased in absolute value by considering a soil‐dependent ring radius. In particular, nearly negligible differences were detected using a small ring in coarse‐textured soils and a large ring in fine‐textured soils. In the coarser soils, inserting the ring and establishing a ponded depth of water did not alter the estimated coefficients of the two‐parameter infiltration model appreciably with the cumulative linearization method, because these coefficients differed between the theoretical and practical setups by no more than 9.2%. In fine soils, linearization could not be possible regardless of the considered setup, or it was the use of d = H = 1 cm instead of d = H = 0 cm that impeded a convincing linearization of the data. In conclusion, the good correspondence, in many circumstances, between the theoretical and the practical Beerkan infiltration experiment reinforced the interest in this simple experiment as a practical means to collect infiltration data in the field.