Multi-mission satellite remote sensing data for improving land hydrological models via data assimilation.

Satellite remote sensing offers valuable tools to study Earth and hydrological processes and improve land surface models. This is essential to improve the quality of model predictions, which are affected by various factors such as erroneous input data, the uncertainty of model forcings, and parameter uncertainties. Abundant datasets from multi-mission satellite remote sensing during recent years have provided an opportunity to improve not only the model estimates but also model parameters through a parameter estimation process. This study utilises multiple datasets from satellite remote sensing including soil moisture from Soil Moisture and Ocean Salinity Mission and Advanced Microwave Scanning Radiometer Earth Observing System, terrestrial water storage from the Gravity Recovery And Climate Experiment, and leaf area index from Advanced Very-High-Resolution Radiometer to estimate model parameters. This is done using the recently proposed assimilation method, unsupervised weak constrained ensemble Kalman filter (UWCEnKF). UWCEnKF applies a dual scheme to separately update the state and parameters using two interactive EnKF filters followed by a water balance constraint enforcement. The performance of multivariate data assimilation is evaluated against various independent data over different time periods over two different basins including the Murray-Darling and Mississippi basins. Results indicate that simultaneous assimilation of multiple satellite products combined with parameter estimation strongly improves model predictions compared with single satellite products and/or state estimation alone. This improvement is achieved not only during the parameter estimation period ([Formula: see text] 32% groundwater RMSE reduction and soil moisture correlation increase from [Formula: see text] 0.66 to [Formula: see text] 0.85) but also during the forecast period ([Formula: see text] 14% groundwater RMSE reduction and soil moisture correlation increase from [Formula: see text] 0.69 to [Formula: see text] 0.78) due to the effective impacts of the approach on both state and parameters.

The role of prior model calibration on predictions with ensemble Kalman filter.

This paper, based on a real world case study (Limmat aquifer, Switzerland), compares inverse groundwater flow models calibrated with specified numbers of monitoring head locations. These models are updated in real time with the ensemble Kalman filter (EnKF) and the prediction improvement is assessed in relation to the amount of monitoring locations used for calibration and updating. The prediction errors of the models calibrated in transient state are smaller if the amount of monitoring locations used for the calibration is larger. For highly dynamic groundwater flow systems a transient calibration is recommended as a model calibrated in steady state can lead to worse results than a noncalibrated model with a well-chosen uniform conductivity. The model predictions can be improved further with the assimilation of new measurement data from on-line sensors with the EnKF. Within all the studied models the reduction of 1-day hydraulic head prediction error (in terms of mean absolute error [MAE]) with EnKF lies between 31% (assimilation of head data from 5 locations) and 72% (assimilation of head data from 85 locations). The largest prediction improvements are expected for models that were calibrated with only a limited amount of historical information. It is worthwhile to update the model even with few monitoring locations as it seems that the error reduction with EnKF decreases exponentially with the amount of monitoring locations used. These results prove the feasibility of data assimilation with EnKF also for a real world case and show that improved predictions of groundwater levels can be obtained.

Plants embrace a stepchild: the discovery of peptide growth regulators.

Over the past decade, peptides have been added to the collection of signalling molecules in plants. As the impact of peptide hormones in non-plants is enormous, a comparison of plant and non-plant peptide signal molecules at this stage deserves our attention-not only to reveal common and unique features, but also to point to new avenues of future research on plant hormones.

The root epidermis-specific pea gene RH2 is homologous to a pathogenesis-related gene.

Two-dimensional gel electrophoresis of pea root and root hair proteins revealed the existence of at least 10 proteins present at elevated levels in root hairs. One of these, named RH2, was isolated and a partial amino acid sequence was determined from two tryptic peptides. Using this sequence information oligonucleotides were designed to isolate by PCR an RH2 cDNA clone. In situ hybridization studies with this cDNA clone showed that rh2 is not only expressed in root hairs, but also in root epidermal cells lacking these tubular outgrowths. During post-embryonic development the gene is switched on after the transition of protoderm into epidermis and since rh2 is already expressed in a globular pea embryo in the protoderm at the side attached to the suspensor, we conclude that the expression of rh2 is developmentally regulated. At the amino acid level RH2 is 95% homologous to the pea PR protein I49a. These gene encoding I49a is induced in pea pods upon inoculation with the pathogen Fusarium solani [12]. We postulate that rh2 contributes to a constitutive defence barrier in the root epidermis. A similar role has been proposed for chalcone synthase (CHS) and chitinase, pathogenesis-related protein that are also constitutively present in certain epidermal tissues.

Characterization of cDNA for nodulin-75 of soybean: A gene product involved in early stages of root nodule development.

Establishment of a nitrogen-fixing root nodule is accompanied by a developmentally regulated expression of nodulin genes, only some of which, the so-called early nodulin genes, are expressed in stages preceding actual nitrogen fixation. We have isolated soybean cDNA clones representing early nodulin genes and have studied clone pENOD2 in detail. The cDNA insert of this clone hybridizes to nodule-specific RNA of 1200 nucleotides in length. The RNA that was hybrid-selected by the cloned ENOD2 DNA was in vitro translated to produce two nodulins with an apparent M(r) of 75,000, the N-75 nodulins. These two nodulins differ slightly in charge and one does not contain methionine. The amino acid sequence deduced from the DNA sequence shows that proline accounts for 45% of the 240 residues in these nodulins and the sequence contains at least 20 repeating heptapeptide units. The amino acid composition of none of the (hydroxy)proline-rich (glyco)proteins described in plants resembles the composition of the N-75 nodulins, especially with respect to the high glutamic acid and the low serine content. This suggests that the N-75 nodulins belong to a hitherto unidentified class of presumably structural proteins. The genes encoding the N-75 nodulins were found to be expressed in nodule-like structures devoid of intracellular bacteria and infection threads, indicating that these nodulins do not function in the infection process but more likely function in nodule morphogenesis.

Nodulin gene expression during soybean (Glycine max) nodule development.

In vitro translation products of total RNA isolated from soybean nodules at successive stages of nodule development were analyzed by two-dimensional gel electrophoresis. In that way the occurrence of over 20 mRNAs specifically transcribed from nodulin genes was detected. The nodulin genes could be divided into two classes according to the time of expression during nodule development. Class A comprises at least 4 nodulin mRNAs which are found when a globular meristem is present in the root cortex. These class A nodulin genes have a transient expression. Class B nodulin genes are expressed when the formation of a nodule structure has been completed. Bradyrhizobium japonicum nod (+) fix(-)mutants, with large deletions spanning the nif H,DK region, still induced nodules showing normal expression of all nodulin genes, indicating that the nif H,DK region is not involved in the induction of nodulin genes. In nodules induced by Bradyrhizobium japonicum nod (+) fix(-)mutant HS124 the bacteria are rarely released from the infection thread and the few infected cells appear to be collapsed. All class A and class B nodulin genes are expressed in HS124 nodules with the exception of 5 class B genes.

cDNA cloning and developmental expression of pea nodulin genes.

A cDNA library prepared from pea nodule poly(A)(+) RNA was screened by differential hybridization with cDNA probes synthesized from root and nodule RNA respectively. From the cDNA clones that hybridized exclusively with the nodule probe five clones, designated pPsNod 6, 10, 11, 13 and 14 and each containing unique sequences, were further characterized together with one leghemoglobin and one "root-specific" cDNA clone. In vitro translation of RNA selected by the pPsNod clones showed that the corresponding genes encode nodulins with molecular weights ranging from 5 800 to 19 000. During pea root nodule development expression of the five PsNod genes starts more or less concomitantly with the onset of nitrogen fixing activity in the nodules and the time course of appearance and accumulation of the nodulin mRNAs is similar to that of leghemoglobin mRNA. In ineffective pea root nodules expression of the PsNod genes is induced but the final accumulation levels of the mRNAs are markedly reduced to various degrees. The expression of another nodulin gene, designated ENOD2, was followed using a heterologous soybean cDNA clone as probe. In pea root nodules the ENOD2 gene is expressed at least five days before the PsNod and leghemoglobin genes, and in contrast to the PsNod mRNAs the concentration of the ENOD2 mRNA is the same in wild type and fix (-) nodules. The results described suggest that in root nodules several regulatory mechanisms exist which determine the final nodulin mRNA amounts accumulating in the root nodule.

Neuroblastoma cells express c-sis and produce a transforming growth factor antigenically related to the platelet-derived growth factor.

Mouse neuroblastoma Neuro-2A cells produce transforming growth factors during exponential growth in a defined hormone-free medium, which, on Bio-Gel columns in 1 M HAc, elute at a molecular size of 15 to 20 kilodaltons (kDa). These neuroblastoma-derived transforming growth factors have strong mitogenic activity, but they do not compete with epidermal growth factor for receptor binding (E. J. J. van Zoelen, D. R. Twardzik, T. M. J. van Oostwaard, P. T. van der Saag, S. W. de Laat, and G. J. Todaro, Proc. Natl. Acad. Sci. U.S.A. 81:4085-4089, 1984). In this study approximately 80% of the mitogenic activity was immunoprecipitated by antibodies raised against platelet-derived growth factor (PDGF). Immunoblotting indicated a true molecular size of 32 kDa for this PDGF-like growth factor. Analysis of poly(A)+ RNA from Neuro-2A cells demonstrated the expression of the c-sis oncogene in this cell line, whereas in vitro translation of the RNA yielded a 20-kDa protein recognized by anti-PDGF antibodies. Separation by reverse-phase high-pressure liquid chromatography demonstrated the presence of two distinct mitogenic activities in neuroblastoma-derived transforming growth factor preparations, one of which is antigenically related to PDGF. Both activities had the ability to induce anchorage-independent growth in normal rat kidney cells, both in the presence and in the absence of epidermal growth factor. It is concluded that Neuro-2A cells express c-sis with concomitant production and secretion of a PDGF-like growth factor, which plays a role in the induction of phenotypic transformation on normal rat kidney cells.

Blotting of RNA onto ion exchange paper allowing subsequent characterization by in situ translation in addition to blot hybridization.

We present the preparation of an ion exchange paper which can be used as a solid carrier in the transfer of RNA from gels. In addition to detection by blot hybridization to specific probes, transferred RNA can be characterized by cell-free translation in situ. Preparation of the paper is simple, inexpensive and reproducible. Examples of applications are shown and possible other applications are discussed.