Productivity throughout the North American Great Plains grasslands is generally considered to be water limited, with the strength of this limitation increasing as precipitation decreases. We hypothesize that cumulative actual evapotranspiration water loss (AET) from April to July is the precipitation‐related variable most correlated to aboveground net primary production (ANPP) in the U.S. Great Plains (GP). We tested this by evaluating the relationship ofANPPtoAET, precipitation, and plant transpiration (Tr). We used multi‐yearANPPdata from five sites ranging from semiarid grasslands in Colorado and Wyoming to mesic grasslands in Nebraska and Kansas, mean annualNRCS ANPP, and satellite‐derived normalized difference vegetation index (NDVI) data. Results from the five sites showed that cumulative April‐to‐JulyAET, precipitation, and Tr were well correlated (R²: 0.54–0.70) to annual changes inANPPfor all but the wettest site.AETand Tr were better correlated to annual changes inANPPcompared to precipitation for the drier sites, and precipitation in August and September had little impact on productivity in drier sites. April‐to‐July cumulative precipitation was best correlated (R² = 0.63) with interannual variability inANPPin the most mesic site, whileAETand Tr were poorly correlated withANPPat this site. Cumulative growing season (May‐to‐September)NDVI(iNDVI) was strongly correlated with annualANPPat the five sites (R² = 0.90). UsingiNDVIas a surrogate forANPP, we found that county‐level cumulative April–JulyAETwas more strongly correlated toANPPthan precipitation for more than 80% of theGPcounties, with precipitation tending to perform better in the eastern more mesic portion of theGP. Including the ratio ofAETto potential evapotranspiration (PET) improved the correlation ofAETto bothiNDVIand mean county‐levelNRCS ANPP. Accounting for how different precipitation‐related variables controlANPP(AETin drier portion, precipitation in wetter portion) provides opportunity to develop spatially explicit forecasting ofANPPacross theGPfor enhancing decision‐making by land managers and use of grasslandANPPfor biofuels.

Every spring, ranchers in the drought‐prone U.S. Great Plains face the same difficult challenge—trying to estimate how much forage will be available for livestock to graze during the upcoming summer grazing season. To reduce this uncertainty in predicting forage availability, we developed an innovative new grassland productivity forecast system, named Grass‐Cast, to provide science‐informed estimates of growing season aboveground net primary production (ANPP). Grass‐Cast uses over 30 yr of historical data including weather and the satellite‐derived normalized vegetation difference index (NDVI)—combined with ecosystem modeling and seasonal precipitation forecasts—to predict if rangelands in individual counties are likely to produce below‐normal, near‐normal, or above‐normal amounts of grass biomass (lbs/ac). Grass‐Cast also provides a view of rangeland productivity in the broader region, to assist in larger‐scale decision‐making—such as where forage resources for grazing might be more plentiful if a rancher’s own region is at risk of drought. Grass‐Cast is updated approximately every two weeks from April through July. Each Grass‐Cast forecast provides three scenarios of ANPP for the upcoming growing season based on different precipitation outlooks. Near real‐time 8‐d NDVI can be used to supplement Grass‐Cast in predicting cumulative growing season NDVI and ANPP starting in mid‐April for the Southern Great Plains and mid‐May to early June for the Central and Northern Great Plains. Here, we present the scientific basis and methods for Grass‐Cast along with the county‐level production forecasts from 2017 and 2018 for ten states in the U.S. Great Plains. The correlation between early growing season forecasts and the end‐of‐growing season ANPP estimate is >50% by late May or early June. In a retrospective evaluation, we compared Grass‐Cast end‐of‐growing season ANPP results to an independent dataset and found that the two agreed 69% of the time over a 20‐yr period. Although some predictive tools exist for forecasting upcoming growing season conditions, none predict actual productivity for the entire Great Plains. The Grass‐Cast system could be adapted to predict grassland ANPP outside of the Great Plains or to predict perennial biofuel grass production.

We used long‐term observations of grassland aboveground net plant production (ANPP, 1939–2016), growing seasonal advanced very‐high‐resolution radiometer remote sensing normalized difference vegetation index (NDVI) data (1982–2016), and simulations of actual evapotranspiration (1912–2016) to evaluate the impact of Pacific Decadal Oscillation (PDO) and El Niño–Southern Oscillation (ENSO) sea surface temperature (SST) anomalies on a semiarid grassland in northeastern Colorado. Because ANPP was well correlated (R² = 0.58) to cumulative April to July actual evapotranspiration (iAET) and cumulative growing season NDVI (iNDVI) was well correlated to iAET and ANPP (R² = 0.62 [quadratic model] and 0.59, respectively), we were able to quantify interactions between the long‐duration (15–30 yr) PDO temperature cycles and annual‐duration ENSO SST phases on ANPP. We found that during cold‐phase PDOs, mean ANPP and iNDVI were lower, and the frequency of low ANPP years (drought years) was much higher, compared to warm‐phase PDO years. In addition, ANPP, iNDVI, and iAET were highly variable during the cold‐phase PDOs. When NINO‐3 (ENSO index) values were negative, there was a higher frequency of droughts and lower frequency of wet years regardless of the PDO phase. PDO and NINO‐3 anomalies reinforced each other resulting in a high frequency of above‐normal iAET (52%) and low frequency of drought (20%) when both PDO and NINO‐3 values were positive and the opposite pattern when both PDO and NINO‐3 values were negative (24% frequency of above normal and 48% frequency of drought). Precipitation variability and subsequent ANPP dynamics in this grassland were dampened when PDO and NINO‐3 SSTs had opposing signs. Thus, primary signatures of these SSTs in this semiarid grassland are (1) increased interannual variability in ANPP during cold‐phase PDOs, (2) drought with low ANPP occurring in almost half of those years with negative values of PDO and NINO‐3, and (3) high precipitation and ANPP common in years with positive PDO and NINO‐3 values.

Recent studies have found that solar ultraviolet (UV) radiation significantly shifts the mass loss and nitrogen dynamics of plant litter decomposition in semi‐arid and arid ecosystems. In this study, we examined the role of photodegradation in litter decomposition by using the DayCent‐UV biogeochemical model. DayCent‐UV incorporated the following mechanisms related to UV radiation: (1) direct photolysis, (2) facilitation of microbial decomposition via production of labile materials, and (3) microbial inhibition effects. We also allowed maximum photodecay rate of the structural litter pool to vary with litter's initial lignin fraction in the model. We calibrated DayCent‐UV with observed ecosystem variables (e.g., volumetric soil water content, live biomass, actual evapotranspiration, and net ecosystem exchange), and validated the optimized model with Long‐Term Intersite Decomposition Experiment (LIDET) observations of remaining carbon and nitrogen at three semi‐arid sites in Western United States. DayCent‐UV better simulated the observed linear carbon loss patterns and the persistent net nitrogen mineralization in the 10‐year LIDET experiment at the three sites than the model without UV decomposition. In the DayCent‐UV equilibrium model runs, UV decomposition increased aboveground and belowground plant production, surface net nitrogen mineralization, and surface litter nitrogen pool, but decreased surface litter carbon, soil net nitrogen mineralization, and mineral soil carbon and nitrogen. In addition, UV decomposition had minimal impacts on trace gas emissions and biotic decomposition rates. The model results suggest that the most important ecological impact of photodecay of surface litter in dry grasslands is to increase N mineralization from the surface litter (25%), and decay rates of the surface litter (15%) and decrease the organic soil carbon and nitrogen (5%).

Missing pixels is a common issue in satellite images. Taking Landsat 8 Analysis Ready Data (ARD) Land Surface Temperature (LST) image as an example, the Source-Augmented Partial Convolution v2 model (SAPC2) is developed to reconstruct missing pixels in the target LST image with the assistance of a collocated complete source image. SAPC2 utilizes the partial convolution enabled U-Net as its framework and accommodates the source into the framework by: (1) performing the shared partial convolution on both the source and the target in encoders; and (2) merging the source and the target by using the partial merge layer to create complete skip connection images for the corresponding decoders. The optimized SAPC2 shows superior performance to four baseline models (i.e., SAPC1, SAPC2-OPC, SAPC2-SC, and STS-CNN) in terms of nine validation metrics. For example, the masked MSE of SAPC2 is 7%, 20%, 44%, and 59% lower than that of the four baseline models. On the six scrutinized cases, the repaired target images generated by SAPC2 have the fewest artifacts near the mask boundary and the best recovery of color scales and fine textures compared with the four baseline models.