1. INTRODUCTION
The Advanced Land Observing Satellite 2 (ALOS-2) is a Japanese earth observation satellite launched in 2014. ALOS-2 observes the earth surface with Phased-Array type L-band Synthetic Aperture Radar 2 (PALSAR-2) that has enhanced performance compared to Japanese previous L-band SAR satellites (i.e., ALOS/PALSAR and JERS-1/SAR) in order to further fulfill social and scientific needs. ALOS-2 is currently in routine operation phase, and until now ALOS-2 has been playing important roles for a lot of applications such as quick response and unique information derived from interferometric observations with L-band SAR are especially effective for monitoring of damaged areas due to earthquakes, volcanic activities, and floods and landslides; crustal deformation mapping and forest change mapping. To keep and enhance the applications using ALOS-2 data, JAXA plans to launch a successor satellite to the ALOS-2 in JFY 2020. In the following sections, we describe the concept of the ALOS-2 follow-on mission (ALOS-4; tentative name).

2. CONCEPTS OF ALOS-2 FOLLOW-ON MISSION
The objectives of the ALOS-4 mission are; (1) sophisticated and practical implementation of land deformation and subsidence monitoring with SAR interferometry, and aiming to use data for not only a posterior damage assessment but also a prior warning assessment, (2) improvement of disaster monitoring performance: more frequent, wider coverage, and readiness for a large-scale disaster such as Nankai Trough Earthquake, (3) continuation and enhancement of the other ALOS-2 mission such as environmental monitoring, (4) new applications such as large infrastructure monitoring with time-series INSAR analysis. The above mission is considered based on the governmental policies and the review on requirements from the ALOS-2 users and lessons learned from ALOS-2 operation and applications.
To achieve the mission, the ALOS-4 satellite should observe wider swath while keeping as higher spatial resolution as that of ALOS-2. The same orbit as ALOS-2 is also required so that users can use a combination of ALOS-2 and ALOS-4 data (e.g., interferometry). The successor to PALSAR-2, namely PALSAR-3, is now designed as a active phased array antenna same as PALSAR-2 with on-board digital beam forming processor in order to improve swath width of 3 m resolution Stripmap mode to 200 km. The swath width of 200 km can cover entire earth surface in 14 days (i.e., one orbit cycle; ALOS-2 PALSAR-2 requires four cycles with 3 m resolution mode). Using the same technique, the swath width of ScanSAR mode increase to 700 km, which can cover a large-scale disaster at once. The data rate also significantly increases with increasing swath width, therefore high-capacity data recorder and high-speed data downlink antennas are prepared. The other specifications of the ALOS-2 follow-on mission are shown in Table 1.

Table 1. Characteristics of the ALOS-2 follow-on mission (ALOS-4)

1) Orbit : Same orbit as ALOS-2,
Sun-synchronous sub-recurrent orbit
Altitude 628 km
Inclination angle 97.9 degree
Local sun time 12:00 ± 15 min. at descending
Revisit: 14 day
Orbit control: within ±500 m from the reference orbit
2) Lifetime : 7 years
3) Satellite Mass: Approx. 3 tons
4) Duty Ratio : maximum 50% (approx. 50 min.)
5) Data Recorder : 1 TByte
6) Downlink : Ka-band (16QAM): 3.6/1.8 Gbps
Optical link (date relay): 1.8 Gbps
7) Launch : JFY 2020, H3 launch vehicle
8) Mission Instruments : PALSAR-3 (Phased Array type L-band Synthetic Aperture Radar-3),
SPAISE3 (SPace based AIS Experiment 3)

Chiba University develops Circularly Polarized Synthetic Aperture Radar (SAR) onboard microsatellite (150 kg class) for global land deformation monitoring. This presentation explains the progress of development of Circularly Polarized SAR sensors for flight tests using unmanned aerial vehicle (UAV) and Boeing 737-200 as microsatellite prelaunch experiments, including anechoic chamber experiment for full polarization of Circularly Polarized SAR scattering. The progress of microsatellite development is also introduced including parabolic mesh antenna, deployment system, and RF system.

The classical synthetic aperture radar (SAR) that adopts linear polarization (horizontal (H) and vertical (V) polarization) is known to be an effective way for monitoring the Earth surface. We approach new SAR configuration that adopts circularly polarized antennas on both transmitter and receiver, namely, circularly polarized SAR. The circularly polarized SAR is known as a robust system for polarization mismatch losses caused by the Faraday rotation effect and antenna misalignment. Also, that of dual polarimetric mode have shown the very promising result regarding classification capability, which is known as compact polarimetry.
To implement and study this concept, we developed circularly polarized SAR constructed by field programmable gate array (FPGA), RF transmitter/receiver, PC, and microstrip circularly polarized patch array antennas, developed for unmanned aerial vehicle (UAV). Also, circularly polarized SAR imaging and target decomposition analysis have been conducted by vector-network analyzer-based circularly polarized SAR system with proper polarimetric calibration technique. For applying to complex real world target, we conducted long-term rice monitoring using ground-based circularly polarized SAR.

Observing temporal variations in ice flow velocities on glaciers and ice sheets is very important for understanding changes in the surrounding environment. We applied Differential Interferometric Synthetic Aperture Radar (DInSAR) and offset tracking to SAR data obtained on Antarctic ice sheet and glaciers for estimating the ice flow velocity.

DInSAR can estimate the velocities of gentle flow on ice sheet and upstream area of glaciers with high ground resolution (several meters), but it is inadequate to observe the displacement exceeding one pixel size of SAR image at downstream area of glaciers during the observation period. Offset tracking method is suitable for estimating the fast ice flow, although the displacement image obtained by this method has low ground resolution (more than 100m). Therefore, by combining these two methods selectively according to the ice flow velocity, it allows us to make an ice flow velocity map over wide area of the Antarctic ice sheet.

As a first step to combine the ice flow velocity maps estimated by DInSAR and the offset tracking, we compared these maps including their accuracy on the overlapped region. The study area was the ice sheet and glaciers around Skallen in the southern part of Sôya Coast, East Antarctica. We used 13 pairs of ALOS/PALSAR data and 2 pairs of ALOS-2/PALSAR-2 data, acquired during the period from November 2007 to January 2011 and from April 2014 to May 2015, respectively. GAMMA software was used for these analyses. The accuracy of the ice flow velocity estimated by DInSAR was approximately 0.04 m/day based on the GNSS measurements on the Antarctic ice sheet (Shiramizu et al., 2016). Based on this verification, we also verified the accuracy of the ice flow velocity estimated by offset tracking.

In this study, we will show the comparison result of two ice flow velocity maps obtained by DInSAR and offset tracking and discuss an approach for combining these ice flow velocity maps with the high precision.

Glacier surge is a periodical orders-of magnitude speed-up event during a short active phase, accompanying terminus advance and ice thickness changes. Near the border of Alaska and Yukon, Canada, there are numerous surge-type glaciers, and their behaviors has received a good deal of scientific attentions. To date, the dynamics have been examined at some surge-type glaciers, but there remain some questions about the generation mechanism.
High-quality images of recent satellites have allowed us to capture the evolutions of surging episodes with high temporal resolution. Steele Glacier in the southwest Yukon is one of the recently activated surge-type glaciers after the quiescence of ~50 years. It experienced the last surge in 1965-1967, and the peak speed was about 24 m/d in early summer 1966 (Stanley, 1969). However, the details of the surging evolution remain unclear. Here we examined the spatial and temporal changes in ice speed, ice thickness and moraines associated with the recent event for the first time in ~50 years.
We used ALOS/PALSAR, Landsat-7, Landsat-8, and Sentienl-1A images to derive the ice speed evolution between 2007 and 2016. Although we have no data in late 2011-early 2013 due to the data availability, RADARSAT-2-based velocity data (Waechter, 2013) showed the latest surge initiated in 2011, and the Sentinel-1A-based velocity data showed it terminated in fall 2016. The observed maximum speed was greater than 20 m d^-1 in early summer 2015, whereas the quiescent speed was ~0.4 m d^-1 between 2007 and 2011. The rapid acceleration changed the ice thickness, which is revealed by Terra/ASTER DEMs. In 2006-2011, the ice thickened above the confluence. In 2013-2016, the ice thickened in the middle and downstream region, while it thinned above the confluence.
Based on the ice thickness changes and the moraine movements, the surge started at the confluence of Hodgson and Steele Glaciers. We will discuss the surge mechanism based on the diverse datasets.

Time series PALSAR-2/ScanSAR data were used for detecting early-stage deforestation. The data were taken 9 times/year, following ALOS-2 systematic observation strategy [1], and it covers global areas, including major tropical forest in the world. By using this data, JICA and JAXA launched a service, “JICA-JAXA Forest Early Warning System in the Tropics (JJ-FAST)” in November 2016 [1]. The system is on the web, and is freely accessible from a smartphone or other devices. Decrease of Gamma⁰_HV are observed after deforestation, and current algorithm uses HV polarization data and two data taken in different timing. HH polarization, and time series data will be used for the future operation. Time series data obtained in South America with HH and HV polarizations were used, and compared it to the Landsat data to clarify the ability to detect the early-stage deforestation, where fallen trees were left on the ground. The Sigma⁰_HH value increased by 1.2 dB in areas undergoing the early stages of deforestation. The detection timing is almost same as that using the optical sensor. On the other hand, the Sigma⁰_HV value decreased by 3.2 dB for late-stage deforestation areas, where fallen trees were removed. The detection timing is about a few month after the detection of deforestation by Sigma⁰_HH, or optical sensor. Many errors of the deforestation detection were observed at wet forest areas. Temporal variations of Gamma⁰ were observed for the area, which induces the deforestation detection errors. The variations of Gamma⁰ shows some correlation with precipitation for both HH and HV polarization [3], and flooding, variation of moisture for soil and trees may be the possible cause for the Gamma⁰ variation.

[1] ALOS-2 systematic observation strategy,
http://www.eorc.jaxa.jp/ALOS/en/top/obs_top.htm Accessed February 16, 2017
[2] JJ-FAST, http://www.eorc.jaxa.jp/jjfast/jj_index.html, February 16, 2017
[3] Manabu Watanabe, et al., Multi-temporal Fluctuations in L-band Backscatter from a Japanese Forest, IEEE Trans. Geosci. Remote Sensing, 53(11), 5799-5813, 2015