MRCE/LTER Soil and Microbial Biomass Nitrogen

Short name: 


Data set ID: 



The data reported here are only for tabonuco forest at El Verde. Following plot setup and one year of pre-treatment measurements, Hurricane Hugo struck before the first planned fertilization. Because a year and a half of annual above ground litter inputs of phosphorus were dropped onto the forest floor in the form of leaf litter, we added a debris removal treatment to the original two, fertilized & control.

These data were used to determine the pools of readily available nitrogen in soil solution in debris removal and fertilizer treatments applied to tabonuco forest at El Verde immediately after Hurricane Hugo struck in September 1989. The treatments (complete removal of litter and woody debris, total fertilization, and control). Both ammonium and nitrate were low in the May 1990 samples. All samples had more KCl extractable ammonium in September, but negligible nitrate. Increases in ammonium were greatest in the fertilized and debris removal plots as compared to the control treatment. These results were positively correlated with litterfall production and canopy closure, with faster rates of recovery in the fertilized and debris removal treatments as compared to the control. Soil extractable (readily available) N was negatively correlated with microbial N in the samples collected one year after Hurricane Hugo, indicating microbial immobilization of N. Century modeling results were consistent with our observations and indicated that woody debris increased the input of low-nutrient substrate to the soil thereby stimulating microbial uptake/immobilization. Thus the soil microbes in the control plots outcompeted the trees for limiting N, thereby slowing their recovery.

Date Range: 
1990-05-05 00:00:00 to 1990-09-15 00:00:00

Publication Date: 

1990-05-05 00:00:00



Additional Project roles: 

Name: Eda Melendez-Colom Role: Data Manager
Name: Jess Zimmerman Role: Associated Researcher
Name: Lawrence R. Walker Role: Associated Researcher
Name: Xiaoming Zou Role: Associated Researcher


The following treatments were applied to plots established within replicate blocks of tabonuco forest at El Verde, Puerto Rico within three weeks after a direct strike by Hurricane Hugo in September 1989: L (removal of all litter and woody debris); F (complete fertilization); and C (control). One control plot had a one-time accidental application of fertilizer between the two sample dates, so measurements were switched to a spare control plot in the same block in September – C(W). Treatments were applied to a 20x20 m area, and samples were taken from the core 10x10 m. See LTERDB111 ( complete methods and map. There were 4 blocks established for this experiment, but one (Block III) was young and unlike the other three late-secondary forest blocks, so it was not included in analyses.

Samples were collected from each subplot and are reported by coordinates by DJL and by subplot numbers by others. DJL coded samples by Block (I, II and IV) and Plot (1-5) to compensate for dyslexia while others coded them as Block-Plot-Subplot numbers (e.g., 1-2-1 for Block II, Plot 1, Subplot 1).

Soil samples were collected volumetrically using an 8.5 cm diameter PCV tubes driven to 10 cm depth on 7 May 1990, and (567.5 cm3) ca. 15 September 1990. Entire soil samples were weighed and sieved through a 5 mm mesh screen, and mixed in a large inflated plastic bag. A 100 g subsample was taken and passed through a 2 mm mesh screen. Rocks sieved out were weighed and this weight was subtracted from the total. Roots were removed by hand. After removal of subsamples for KCl extractions, a ca. 20 g subsample was transferred to an aluminum pan, the wet weight recorded, then dried at 50C for at least one week, and the dry weight recorded.

Soil solution nitrogen was extracted from 2 replicate subsamples (ca. 4 g wet weight) in 100 ml 2N KCl within 24 h of collection. KCl suspensions were placed on a wrist-action shaker for 1 hour and then filtered with suction using Whatman #10 filter papers. Soil extracts were stored frozen until analyzed. The samples collected in 15 September 1990 were intended for analysis of microbial N (fumigation-extraction method) as well as measuring the pool of extractable soil nitrate and ammonium in the control (initial) samples. Fumigated samples were placed in a desiccator with chloroform and the head space was evacuated until the chloroform boiled. Fumigated soil was then extracted in 2N KCl, as above. Ammonium-N and nitrate-N (mg/L) in the KCl extracts were determined spectrophotometrically using the phenate and cadmium reduction methods, respectively (Technicon 1985). Some soil extracts were misplaced or lost during power outages, so data from some plots are missing. However, there was at least one replicate plot per treatment of remaining samples. Soil bulk density was calculated from the September samples.



Time Period: