Artificial Gaps Research Articles

Univariate and multivariate imputation methods evaluation for reconstructing climate time series data: A case study of Mosul station-Iraq

Comprehensive climate time series data is indispensable for monitoring the impacts of climate change. However, observational datasets often suffer from data gaps within their time series, necessitating imputation to ensure dataset integrity for further analysis. This study evaluated six univariate and multivariate imputation methods to infill missing values. These methods were applied to complete the subsets of time series data for precipitation, temperature, and relative humidity from Mosul station spanning 1980–2013. Artificial gaps of 5%, 10%, 20%, and 30% missing observations were introduced under scenarios of missing completely at random (MCAR) missing at random (MAR), and missing not at random (MNAR). Evaluation metrics including RMSE and Kling-Gupta Efficiency were utilized for performance evaluation. Results revealed that seasonal decomposition was the most effective univariate imputation method across all variables. For the multivariate imputation, kNN demonstrated superior performance in infilling the precipitation missing data under MCAR, while norm.predict exhibited optimal performance in the temperature missing data under all missing scenarios. Moreover, missForest was identified as the most suitable method for infilling missing relative humidity data. This study's methodology offers insights into selecting appropriate imputation methods for other climate stations, thereby enhancing the accuracy of the climate change effects analysis.

Monsoon-Based Linear Regression Analysis for Filling Data Gaps in Gravity Recovery and Climate Experiment Satellite Observations

Over the past two decades, the Gravity Recovery and Climate Experiment (GRACE) satellite mission and its successor, GRACE-follow on (GRACE-FO), have played a vital role in climate research. However, the absence of certain observations during and between these missions has presented a persistent challenge. Despite numerous studies attempting to address this issue with mathematical and statistical methods, no definitive optimal approach has been established. This study introduces a practical solution using Linear Regression Analysis (LRA) to overcome data gaps in both GRACE data types—mascon and spherical harmonic coefficients (SHCs). The proposed methodology is tailored to monsoon patterns and demonstrates efficacy in filling data gaps. To validate the approach, a global analysis was conducted across eight basins, monitoring changes in total water storage (TWS) using the technique. The results were compared with various geodetic products, including data from the Swarm mission, Institute of Geodesy and Geoinformation (IGG), Quantum Frontiers (QF), and Singular Spectrum Analysis (SSA) coefficients. Artificial data gaps were introduced within GRACE observations for further validation. This research highlights the effectiveness of the monsoon method in comparison to other gap-filling approaches, showing a strong similarity between gap-filling results and GRACE’s SHCs, with an absolute relative error approaching zero. In the mascon approach, the coefficient of determination (R2) exceeded 91% for all months. This study offers a readily usable gap-filling product—SHCs and smoothed gridded observations—with accurate error estimates. These resources are now accessible for a wide range of applications, providing a valuable tool for the scientific community.

A Modified Version of the Direct Sampling Method for Filling Gaps in Landsat 7 and Sentinel 2 Satellite Imagery in the Coastal Area of Rhone River

Earth Observation (EO) data, such as Landsat 7 (L7) and Sentinel 2 (S2) imagery, are often used to monitor the state of natural resources all over the world. However, this type of data tends to suffer from high cloud cover percentages during rainfall/snow seasons. This has led researchers to focus on developing algorithms for filling gaps in optical satellite imagery. The present work proposes two modifications to an existing gap-filling approach known as the Direct Sampling (DS) method. These modifications refer to ensuring the algorithm starts filling unknown pixels (UPs) that have a specified minimum number of known neighbors (Nx) and to reducing the search area to pixels that share similar reflectance as the Nx of the selected UP. Experiments were performed on images acquired from coastal water bodies in France. The validation of the modified gap-filling approach was performed by imposing artificial gaps on originally gap-free images and comparing the simulated images with the real ones. Results indicate that satisfactory performance can be achieved for most spectral bands. Moreover, it appears that the bi-layer (BL) version of the algorithm tends to outperform the uni-layer (UL) version in terms of overall accuracy. For instance, in the case of B04 of an L7 image with a cloud percentage of 27.26%, accuracy values for UL and BL simulations are, respectively, 64.05 and 79.61%. Furthermore, it has been confirmed that the introduced modifications have indeed helped in improving the overall accuracy and in reducing the processing time. As a matter of fact, the implementation of a conditional filling path (minNx = 4) and a targeted search (n2 = 200) when filling cloud gaps in L7 imagery has contributed to an average increase in accuracy of around 35.06% and an average gain in processing time by around 78.18%, respectively.

The application of gap planting technology to rehabilitate degraded natural forest

In 2020 degraded forests in Indonesia reached 33.4 million ha. These forests must be rehabilitated. The purpose of this research was to show the application of gap planting technology to rehabilitate degraded natural forests. This improvement of forest regeneration methods by planting local intolerant tree species in gaps area was used. Artificial gaps area was designed and it’s location was adjusted to the topography and forest cover. The size of each gap area is 2,000 m2. The ratio of total gaps area to total natural forests area was about four to six. Open areas, soil density, forest fire risk, working time and cost of gap planting, and tree growth were studied. The results showed that after the gap planting: forest cover consists of 37.43% gap planting area and 62.57% natural forest area; forest soil is not compacted; the fire hazard rating in the gap and natural forests is low. Moreover, it is cost-effective and has positive impacts such as job creation for local communities and a productivity increase of rehabilitated forests. It is concluded that the gap planting technology was environment friendly technology, and it is recommended to be used for rehabilitation of degraded forests in Indonesia.

Disentangling regeneration by vertical stratification: A 17-year gap-filling process in a temperate secondary forest

Forest gaps are widely spread in the forest ecosystems and play important roles in driving forest dynamics. However, long-term regeneration during the gap-filling process was less understood, especially species composition and dynamics along vertical gradients over time. We monitored tree regeneration in artificial gaps of a temperate secondary forest for 17 years and analyzed the gap-filling process both as a whole and by stratification (lower layer: 0∼3 m, middle layer: 3∼6 m, and upper layer: >6 m) for three shade-tolerance groups of tree species. We found that the interaction effects of gap size and shade tolerance could result in a growth difference of five to ten years among different regeneration groups, with the most significant effects in the large gaps for the light-demanding species such as Juglans mandshurica and Phellodendron amurense. Regeneration density in the large and small gaps eventually converged, but there were great differences in density dynamics. The regeneration density in the large gaps decreased gradually over time, while it decreased sharply in the small gaps after just one growing season of gap formation and consistently remained at low levels regardless of shade tolerance. Species richness decreased smoothly during the 17-year monitoring, but the richness in the large gaps was consistently higher than that in the small gaps, with all shade-tolerance groups existing, which underlined the importance of gap size in maintaining tree species diversity. Differences in the density, richness, and composition among regeneration layers reduced over time. However, vertical regeneration patterns diverged among shade-tolerance groups during the gap-filling process. Light-demanding species were likely to have only one chance for successful regeneration because almost no new recruitments appeared in the lower or middle layer as dominant stems reached the upper layer. Intermediate and shade-tolerant species, by contrast, were well-distributed in different regeneration layers, indicating a stronger capacity for resistance and resilience during uncertain disturbance events in the long run. Our results revealed the gap-filling strategies of tree species with different shade-tolerance groups, which explained why light-demanding species could not completely occupy a forest gap and highlighted the necessity of repeated gap disturbance for intermediate and shade-tolerant species regeneration.

Preenchimento de falhas em dados meteorológicos usando o programa GapMET no Estado de Mato Grosso, Brasil

ABSTRACT This paper aimed to introduce the GapMET software, developed by the authors, and evaluate the accuracy of its six methods for gap-filling the main meteorological variables monitored by weather station in the state of Mato Grosso, Brazil, using reference time series from neighbour weather station and/or remote sensing products. The methods were tested on seven different databases, with 25 to 80% artificial gaps, and their accuracy was given by the number of gaps left unfilled, the bias, the RMSE, and Pearson’s correlation. The GapMET software showed good results in filling meteorological gaps regardless of the method applied. Methods that use only one neighbour weather station as a reference series showed better results because, in the state, the minimum distance for a weather station to have at least three neighbours as reference was 350 km, reducing the climatic similarity between them and consequently the accuracy when more than one reference series were needed. The use of satellite reference series reduced the probability of unfilled gaps; however, it showed higher bias and RMSE and lower correlations.

The distribution of the tree of heaven (Ailanthus altissima (Mill.) Swingle) in the settlements and forests of Southern Börzsöny, Hungary

: In this study, we investigated the occurrence of the invasive alien tree of heaven (Ailanthus altissima) in the approximately 2,100-hectare area of Southern Börzsöny, along the road network in settlements Zebegény and Nagymaros and in forests. Th e survey was carried out according to the abundance and location of the species (for all occurrences), the maintained or neglected status of the area (for settlements), and the open or closed nature of the stand (for forests). Individuals or more coherent stands of the tree of heaven were recorded using handheld GPS. A new point was taken when the distance between two individuals or stands exceeded 10 m. We recorded 482 occurrences of tree of heaven in the two settlements and found that it occurs most frequently in neglected public green areas (38%) and neglected gardens (21%), and oft en shows up along railway lines (18%). In forested areas between the two settlements, we recorded 193 occurrences and found that the tree occurs mainly along regularly used forest roads (42%), and most frequently in open stands (especially in artifi cial gaps) (69%). Our study also points out that the Hungarian National Forest Inventory data indicate signifi cantly fewer occurrences (covering 23 subcompartments) than our forest mapping survey or earlier ones by other researchers (56 and 60 subcompartments, respectively). Based on our results, anthropogenic impacts are crucial for the spread of the species. Th e tree is expected to spread exponentially in the future, so in order to prevent this, we recommend that treatments are carried out as soon as possible, with the immediate eradication of seed-producing individuals. Th e issue of social perception and knowledge is also crucial, as the majority of the society generally has little knowledge of both the tree of heaven and the dangers posed by invasive plant species. Targeted education of the public can therefore be an important element of prevention.

Self-supervised learning-based Multi-Scale feature Fusion Network for survival analysis from whole slide images

Understanding prognosis and mortality is critical for evaluating the treatment plan of patients. Advances in digital pathology and deep learning techniques have made it practical to perform survival analysis in whole slide images (WSIs). Current methods are usually based on a multi-stage framework which includes patch sampling, feature extraction and prediction. However, the random patch sampling strategy is highly unstable and prone to sampling non-ROI. Feature extraction typically relies on hand-crafted features or convolutional neural networks (CNNs) pre-trained on ImageNet, while the artificial error or domain gaps may affect the survival prediction performance. Besides, the limited information representation of local sampling patches will create a bottleneck limitation on the effectiveness of prediction. To address the above challenges, we propose a novel patch sampling strategy based on image information entropy and construct a Multi-Scale feature Fusion Network (MSFN) based on self-supervised feature extractor. Specifically, we adopt image information entropy as a criterion to select representative sampling patches, thereby avoiding the noise interference caused by random to blank regions. Meanwhile, we pretrain the feature extractor utilizing self-supervised learning mechanism to improve the efficiency of feature extraction. Furthermore, a global–local feature fusion prediction network based on the attention mechanism is constructed to improve the survival prediction effect of WSIs with comprehensive multi-scale information representation. The proposed method is validated by adequate experiments and achieves competitive results on both of the most popular WSIs survival analysis datasets, TCGA-GBM and TCGA-LUSC. Code and trained models are made available at: https://github.com/Mercuriiio/MSFN.

Creating an anatomical wax-up in partially edentulous patients by means of a statistical shape model.

Creating wax-ups of missing teeth for backward planning in implant surgery is a complex and time-consuming process. To facilitate implant-planning procedures, the automatic generation of a virtual wax-up would be useful. In the present study, the reconstruction of missing teeth in partially edentulous patients was performed automatically using newly developed software. The accuracy was investigated in order to test its clinical applicability. This study presents a new method for creating an automatic virtual wax-up, which could serve as a basic tool in modern implant-planning procedures. First, a statistical shape model (SSM) based on 76 maxillary and mandibular arch scans from dentally healthy individuals was generated. Then, artificially generated tooth gaps were reconstructed. The accuracy of the workflow was evaluated on a separate testing sample of 10 individuals with artificially created tooth gaps given as a median deviation, in millimeters. Scans of three clinical cases with partial edentulism were equally reconstructed using the SSM and compared with the final prosthodontic work. The reconstruction of the artificial tooth gaps could be performed with the following median reconstruction accuracy: gap 21 with 0.15 mm; gap 27 with 0.20 mm; gap 34 with 0.22 mm: gap 36 with 0.22 mm; gaps 12 to 22 with 0.22 mm; gaps 34 to 36 with 0.22 mm. A scenario for an almost edentulous mandible with all teeth missing except teeth 33 and 43 could be reconstructed with a median reconstruction accuracy of 0.37 mm. The median tooth gap deviation of the SSM-based reconstruction in clinical cases differed from the final inserted prosthodontic teeth by 0.49 to 0.86 mm in median. A first feasibility of creating virtual wax-ups using an SSM could be shown. Artificially generated tooth gaps could be reconstructed close to the original with the proposed workflow. In the clinical cases, the SSM proposes an anatomical reconstruction, which does not yet consider prosthodontic aspects. To obtain clinical use, contact with antagonist teeth must be considered and more training data must be implemented. However, the presented method offers a fast and viable way for the approximate placement of missing crowns. This could be used in a digital planning workflow when implant position must be determined. (Int J Comput Dent 2022;25(4):349-0; doi: 10.3290/j.ijcd.b2599407).

Investigation of scarce input data augmentation for modelling nitrogenous compounds in South African rivers

Abstract In this study, basic interpolation and machine learning data augmentation were applied to scarce data used in Water Quality Analysis Simulation Programme (WASP) and Continuous Stirred Tank Reactor (CSTR) that were applied to nitrogenous compound degradation modelling in a river reach. Model outputs were assessed for statistically significant differences. Furthermore, artificial data gaps were introduced into the input data to study the limitations of each augmentation method. The Python Data Analysis Library (Pandas) was used to perform the deterministic interpolation. In addition, the effect of missing data at local maxima was investigated. The results showed little statistical difference between deterministic interpolation methods for data augmentation but larger differences when the input data were infilled specifically at locations where extrema occurred.

Multiple gap-filling for eddy covariance datasets

With novel developments in technology, eddy covariance flux measurements have become feasible for a variety of trace gases. While the statistical properties and gap-filling strategies have been well examined for carbon dioxide, these are much less understood for other gases.Here, we propose a universal methodology deploying multiple gap-filling techniques and artificial gap scenarios to evaluate the techniques’ performances, infer the statistical flux properties, and fill the real gaps in eddy covariance datasets of any trace gas. The methodology was implemented in a gap-filling framework with techniques spanning from simple and diurnal interpolations, look-up tables, artificial neural networks, to an inferential model. For the new scheme of half-hourly and daily artificial gaps, each additional gap was superimposed one at a time (thus keeping the disturbance to a minimum) for the whole dataset and the scenarios were resampled by bootstrapping. The gap-filled sums were then estimated from the ensemble of well-performing gap-filling techniques. The gap-filling framework was applied to campaign data of three different trace gases (51 days of ammonia, 79 days of total reactive nitrogen, and 89 days of methane flux measurements). The aggregated fluxes are stated as ensemble ranges of multiple techniques plus the techniques’ uncertainties. Additionally, the framework was used to gap-fill a full year of carbon dioxide flux measurements yielding similar performances as previously reported.Based on a review of gap-filling comparison studies and on our findings, we suggest reconsidering the standard procedure of using one gap-filling technique for multi-site studies. Deploying multiple gap-filling techniques and providing ensemble results of gap-filled sums will help to minimize the influence of a single technique and thus lead to a more robust flux aggregation. Furthermore, the estimated overall uncertainty will be more realistic by accounting for the ensemble range of multiple techniques.

Exploring the Potential of Neural Networks to Predict Statistics of Solar Wind Turbulence

AbstractTime series data sets often have missing or corrupted entries, which need to be handled in subsequent data analysis. For example, in the context of space physics, calibration issues, satellite telemetry issues, and unexpected events can make parts of a time series unusable. This causes problems for understanding the dynamics of the heliosphere and space weather environment. Various approaches exist to tackle this problem, including mean/median imputation, linear interpolation, and autoregressive modeling. Here, we study the utility of artificial neural networks (ANNs) to predict statistics of sparse time series. Our focus is not on time series prediction but on gleaning the best possible information about the statistical behavior of the system. As an example application, we focus on the structure functions of turbulent time series measured in the solar wind. Using a data set with artificial gaps, a neural network is trained to predict second‐order structure functions and then tested on an unseen data set to quantify its performance. A small feedforward ANN, with only 20 hidden neurons, can predict the large‐scale fluctuation amplitudes better than mean imputation or linear interpolation when the percentage of missing data is high. Although they perform worse than the other methods when it comes to capturing both the shape and fluctuation amplitude together, their performance is better in a statistical sense for large fractions of missing data. Caveats regarding their utility, the optimization procedure, and potential future improvements are discussed.

Estimating gap age using tree-ring width in combination with carbon isotope discrimination in a temperate forest, Northeast China

Key messageThe accuracy of gap age estimation can be improved from 5–6 to 2 years by analyzing the annual width and carbon isotope discrimination of the rings of gap-surrounding trees.ContextGap age has a direct link to the composition and structure of regeneration. However, the accuracy of gap age estimation is still limited.AimWe aim to improve the accuracy of gap age estimation by analyzing the width in combination with carbon isotope discrimination (∆13C) of the rings of gap-surrounding trees.MethodsTwenty-four gap-surrounding trees (nine Ulmus laciniata, eight Fraxinus rhynchophylla, and seven Juglans mandshurica) were selected from eight artificial gaps created in December 2004. First, the growth release (i.e., peak time of percent growth change) for sample trees was measured based on the tree-ring width to identify the rough time range of gap formation. Then, the ∆13C of rings during the time range were analyzed for determining the precise year of gap formation.ResultsThe peak time of percent growth change occurred from 2005 to 2010 for U. laciniata, 2004 to 2008 for F. rhynchophylla, and 2002 to 2007 for J. mandshurica. Within the range of 2002 to 2010, the ∆13C of rings for all sample trees significantly reduced in 2005–2006 (p < 0.05), which was the estimated year of gap formation.ConclusionThe introduction of ∆13C analysis could effectively reduce the estimating deviations of gap age by only considering tree-ring width analyses and finally improve the accuracy of gap age estimation within 2 years, which can provide reliable information for gap management.

Assessing soil organic carbon, N and P stocks and its relation to soil properties in artificial canopy gaps in a managed oriental beech (Fagus orientalis L.) forest#

AbstractBackgroundIn the Hyrcanian temperate forests, different canopy gaps sizes have been created by logging under silvicultural methods. Understanding the effects of gaps on forest ecosystems can help in the planning and decision‐making of forestry practices. Variations in ecological conditions caused by canopy gaps may have an impact on the forest stand, particularly soil nutrients.AimsThe purpose of the present research was to investigate the changes in soil organic carbon (SOC), total nitrogen (N) and phosphorus (P) stocks under different gap sizes in the Hyrcanian oriental beech forest.MethodsWe measured SOC, N and P stocks of soil at the depth of 0–20 cm among 15 artificial gaps along with adjacent closed stand which was created by single‐tree selection system in oriental beech stand, northern Iran. In the center and four cardinal positions of the edge of gaps as well as in the four adjacent closed canopy forests at the distance of 20 m from gaps, the soil sampling was performed.ResultsThe difference in C, N and P stocks among the artificial canopy gaps was statistically significant, and higher mean values were observed in large gaps. No significant difference was observed between gaps and adjacent closed stands with regard to soil C, N and P stocks. At artificial gaps and closed stands, the SOC stock was negatively correlated with clay and positively correlated with bulk density, sand, N and OC. With increasing bulk density, sand, N, P, and OC, N stock was increased in gaps. In gaps, SOC and P stocks were both positively related to soil microbial C and soil microbial P, respectively.ConclusionGenerally, it can be stated that there are similar relative ecological conditions between artificial gaps and closed stands, and the interaction among soil properties in gaps demonstrates gap dynamics in deciduous, broad‐leaved beech forest.

A Novel Framework for Unification of Association Rule Mining, Online Analytical Processing and Statistical Reasoning

Statistical reasoning was one of the earliest methods to draw insights from data. However, over the last three decades, association rule mining and online analytical processing have gained massive ground in practice and theory. Logically, both association rule mining and online analytical processing have some common objectives, but they have been introduced with their own set of mathematical formalizations and have developed their specific terminologies. Therefore, it is difficult to reuse results from one domain in another. Furthermore, it is not easy to unlock the potential of statistical results in their application scenarios. The target of this paper is to bridge the artificial gaps between association rule mining, online analytical processing and statistical reasoning. We first provide an elaboration of the semantic correspondences between their foundations, i.e., itemset apparatus, relational algebra and probability theory. Subsequently, we propose a novel framework for the unification of association rule mining, online analytical processing and statistical reasoning. Additionally, an instance of the proposed framework is developed by implementing a sample decision support tool. The tool is compared with a state-of-the-art decision support tool and evaluated by a series of experiments using two real data sets and one synthetic data set. The results of the tool validate the framework for the unified usage of association rule mining, online analytical processing, and statistical reasoning. The tool clarifies in how far the operations of association rule mining and online analytical processing can complement each other in understanding data, data visualization and decision making.

Spectral Temporal Information for Missing Data Reconstruction (STIMDR) of Landsat Reflectance Time Series

The number of Landsat time-series applications has grown substantially because of its approximately 50-year history and relatively high spatial resolution for observing long term changes in the Earth’s surface. However, missing observations (i.e., gaps) caused by clouds and cloud shadows, orbit and sensing geometry, and sensor issues have broadly limited the development of Landsat time-series applications. Due to the large area and temporal and spatial irregularity of time-series gaps, it is difficult to find an efficient and highly precise method to fill them. The Missing Observation Prediction based on Spectral-Temporal Metrics (MOPSTM) method has been proposed and delivered good performance in filling large-area gaps of single-date Landsat images. However, it can be less practical for a time series longer than one year due to the lack of mechanics that exclude dissimilar data in time series (e.g., different phenology or changes in land cover). To solve this problem, this study proposes a new gap-filling method, Spectral Temporal Information for Missing Data Reconstruction (STIMDR), and examines its performance in Landsat reflectance time series. Two groups of experiments, including 2000 × 2000 pixel Landsat single-date images and Landsat time series acquired from four sites (Kenya, Finland, Germany, and China), were performed to test the new method. We simulated artificial gaps to evaluate predicted pixel values with real observations. Quantitative and qualitative evaluations of gap-filled images through comparisons with other state-of-the-art methods confirmed the more robust and accurate performance of the proposed method. In addition, the proposed method was also able to fill gaps contaminated by extreme cloud cover for a period (e.g., winter in high-latitude areas). A down-stream task of random forest supervised classification through both gap-filled simulated datasets and the original valid datasets verified that STIMDR-generated products are relevant to the user community for land cover applications.

Stable gap-filling for longer eddy covariance data gaps: A globally validated machine-learning approach for carbon dioxide, water, and energy fluxes

Continuous time-series of CO2, water, and energy fluxes are useful for evaluating the impacts of climate-change and management on ecosystems. The eddy covariance (EC) technique can provide continuous, direct measurements of ecosystem fluxes, but to achieve this gaps in data must be filled. Research-standard methods of gap-filling fluxes have tended to focus on CO2 fluxes in temperate forests and relatively short gaps of less than two weeks. A gap-filling method applicable to other fluxes and capable of filling longer gaps is needed.To address this challenge, we propose a novel gap-filling approach, Random Forest Robust (RFR). RFR can accommodate a wide range of data gap sizes, multiple flux types (i.e. CO2, water and energy fluxes). We configured RFR using either three (RFR3) or ten (RFR10) driving variables. RFR was tested globally on fluxes of CO2, latent heat (LE), and sensible heat (H) from 94 suitable FLUXNET2015 sites by using artificial gaps (from 1 to 30 days in length) and benchmarked against the standard marginal distribution sampling (MDS) method.In general, RFR improved on MDS's R2 by 15% (RFR3) and by 30% (RFR10) and reduced uncertainty by 70%. RFR's improvements in R2 for H and LE were more than twice the improvement observed for CO2 fluxes. Unlike MDS, RFR performed well for longer gaps; for example, the R2 of RFR methods in filling 30-day gaps dropped less than 4% relative to 1-day gaps, while the R2 of MDS dropped by 21%.Our results indicate that the RFR method can provide improved gap-filling of CO2, H and LE flux timeseries. Such improved continuous flux measurements, with low bias, can enhance our understanding of the impacts of climate-change and management on ecosystems globally.

Effective method for filling gaps in time series of environmental remote sensing data: An example on evapotranspiration and land surface temperature images

Phenomena such as cloudiness, atmospheric aerosol or sensor failure cause missing data (gaps) in remote sensing images and damage spatial and temporal continuity. Dealing with this deficiency is of importance for continuous spatio-temporal modeling and environmental studies. Simplicity, efficiency and accuracy are dominant factors in practicality of gap filling algorithms especially in dealing with large gaps and long time series. In this study, an effective and efficient spatio-temporal gap filling algorithm is proposed, implemented and tested. The method was applied to MODIS 8-day Land Surface Temperature (LST) and Evapotranspiration (ET) datasets with a 1 km spatial resolution. To assess the performance of the proposed methods, artificial gaps were introduced and filled. Then estimated and real values were compared. The results showed that our method can predict the missing values very accurately (based on RMSE) even in gaps with heterogeneous surface for both variables. Our proposed method has no limitation on the shape and size of the gaps. The proposed algorithm is flexible regarding parameterization. It can handle large volumes of datasets due to its parallelizable structure. More importantly, the method run-time is extremely low even in large gaps.

Technical note: Uncertainties in eddy covariance CO&amp;lt;sub&amp;gt;2&amp;lt;/sub&amp;gt; fluxes in a semiarid sagebrush ecosystem caused by gap-filling approaches

Abstract. Gap-filling eddy covariance CO2 fluxes is challenging at dryland sites due to small CO2 fluxes. Here, four machine learning (ML) algorithms including artificial neural network (ANN), k-nearest neighbors (KNNs), random forest (RF), and support vector machine (SVM) are employed and evaluated for gap-filling CO2 fluxes over a semiarid sagebrush ecosystem with different lengths of artificial gaps. The ANN and RF algorithms outperform the KNN and SVM in filling gaps ranging from hours to days, with the RF being more time efficient than the ANN. Performances of the ANN and RF are largely degraded for extremely long gaps of 2 months. In addition, our results suggest that there is no need to fill the daytime and nighttime net ecosystem exchange (NEE) gaps separately when using the ANN and RF. With the ANN and RF, the gap-filling-induced uncertainties in the annual NEE at this site are estimated to be within 16 g C m−2, whereas the uncertainties by the KNN and SVM can be as large as 27 g C m−2. To better fill extremely long gaps of a few months, we test a two-layer gap-filling framework based on the RF. With this framework, the model performance is improved significantly, especially for the nighttime data. Therefore, this approach provides an alternative in filling extremely long gaps to characterize annual carbon budgets and interannual variability in dryland ecosystems.

Determining Temporal Uncertainty of a Global Inland Surface Water Time Series

Earth observation time series are well suited to monitor global surface dynamics. However, data products that are aimed at assessing large-area dynamics with a high temporal resolution often face various error sources (e.g., retrieval errors, sampling errors) in their acquisition chain. Addressing uncertainties in a spatiotemporal consistent manner is challenging, as extensive high-quality validation data is typically scarce. Here we propose a new method that utilizes time series inherent information to assess the temporal interpolation uncertainty of time series datasets. For this, we utilized data from the DLR-DFD Global WaterPack (GWP), which provides daily information on global inland surface water. As the time series is primarily based on optical MODIS (Moderate Resolution Imaging Spectroradiometer) images, the requirement of data gap interpolation due to clouds constitutes the main uncertainty source of the product. With a focus on different temporal and spatial characteristics of surface water dynamics, seven auxiliary layers were derived. Each layer provides probability and reliability estimates regarding water observations at pixel-level. This enables the quantification of uncertainty corresponding to the full spatiotemporal range of the product. Furthermore, the ability of temporal layers to approximate unknown pixel states was evaluated for stratified artificial gaps, which were introduced into the original time series of four climatologic diverse test regions. Results show that uncertainty is quantified accurately (&gt;90%), consequently enhancing the product’s quality with respect to its use for modeling and the geoscientific community.